€¦ · web viewbs s = c s w v a . where bed sediment storage (bs. s), reported as grams per...
TRANSCRIPT
Highly conservative behaviour of bed sediment-associated
metals following extreme flooding
Rachel R. Hurleya*, Jamie C. Woodwarda & James J. Rothwella
a Department of Geography, The University of Manchester, Manchester, UK, M13 9PL
1
1
2
3
4
5
6
7
8
9
Abstract
On 26th December (Boxing Day) 2015 an exceptional flood event occurred in the Irwell
catchment, UK, whilst the neighbouring Mersey catchment experienced a much more typical
winter runoff event. This provided an opportunity to examine the influence of high magnitude
hydrological processes on the behaviour of fine-grained metal-contaminated bed sediments.
Forty sites across the two catchments were sampled for channel bed fine sediment storage
and sediment-associated metal(loid) concentrations prior to, and following, the flooding.
Sediments were analysed for total As, Cr, Cu, Pb, and Zn, and then subjected to a 5-step
sequential extraction procedure. Despite a significant reorganisation of fine-grained (<63 µm)
sediment storage, metal(loid) concentrations demonstrated markedly conservative behaviour
with no significant difference observed between pre- and post-flooding values across both
catchments. Estimates of the channel bed storage of sediment-associated metal(loid)s also
showed minimal change as a result of the flooding. The metal partitioning data reveal only
minor changes in the mobility of bed sediment-associated metal(loid)s, indicating that such
flood events do not increase the availability of sorbed contaminants in these catchments.
Post-flooding bed sediment metal(loid) loadings remain high, indicating persistent and long-
lasting sources of contamination within the Irwell and upper Mersey fluvial network.
Keywords: metal(loid)s, bed sediment, river, flooding, sediment storage, metal storage,
sequential extraction, hydrology
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
1. Introduction
Exceptional flood events have been associated with the greatest riverine loads of particulate
metal export (Peraza-Castro, Sauvage, Sánchez-Pérez & Ruiz-Romera, 2016). Over 90% of a
river’s annual suspended sediment flux may be associated with storm events (Walling, Webb
& Woodward, 1992), which can account for >90% of trace metal transport (Horowitz, Elrick
& Smith, 2008). Moreover, in flashy, urbanised catchments, exceptional individual events,
which may represent only 1% of the hydrological year, can transport >40% of the annual
suspended sediment load (Old et al., 2006). Such large events deliver high volumes of
sediment and contaminants downstream and have the potential to significantly reorganise
spatial patterns of sediment storage and associated metals within river catchments.
Significant variability in sediment-associated contaminant transport during flood events has
been reported in a range of catchment types (e.g. Leenaers, 1989a; Zonta, Collavini, Zaggia
& Zuliani, 2005; Coynel, Schäfer, Blanc & Bossy, 2007; Resongles, Casiot & Freydier,
2015). This includes varying concentrations across the flood hydrograph, where elevated
concentrations are often dependent on discharge (Bradley and Lewin, 1982; Zonta et al.,
2005) and contaminants may exhibit a ‘first flush’ response (Lee, Bang, Ketchum & Choe,
2002; Yin and Li, 2008). Changing hydrological conditions within a flood event have been
shown to access different catchment sediment sources, leading to changes in sediment-
associated contaminant concentrations and other sediment characteristics (Walling and
Woodward, 1992; 1995; Rothwell, Robinson, Evans, Yang & Allott, 2005). A dilution effect
is sometimes observed at peak discharge and during the falling limb of the flood hydrograph
as contaminated sediment sources are exhausted and clean material is introduced into the
fluvial system from uncontaminated headwater environments (e.g. Bradley and Lewin, 1982;
Bradley, 1984; Dawson and Macklin, 1998a; Coynel et al., 2007). However, urbanisation and
industrialisation can influence the dynamics of metal cycling during and following flooding.
3
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
These factors may influence catchment hydrological processes and the redistribution of
metals, so that sediment-associated metal concentrations largely reflect spatial sources (e.g.
Hutchinson and Rothwell, 2008; Baborowski and Einax, 2016).
Studies of bed sediment-associated metals demonstrate significant change in response to
flood events. For example, in mining-affected catchments, concentrations associated with
channel bed sediments typically decrease downstream, albeit with some small increases
observed close to the mining site (Ciszewski, 2001; Wadige, Taylor, Krikowa & Maher,
2016). In contrast, Miller et al. (1999) found no change in mine tailing-derived Hg
concentrations within bed sediments of the Carson River Valley, USA, despite dramatic shifts
in channel morphology. Metal concentrations within the fine-grained channel bed sediments
of urbanised catchments have been shown to persist across multiple flood events and flow
conditions (Horowitz, Elrick, Smith & Stephens, 2014; Pulley, Foster & Antunes, 2016),
reflecting the continuity of catchment sources. These studies included contributions from
urban road deposited sediments (RDS) that were not depleted by successive storm events,
although this may reflect a replenishment of contaminant stores between events (Pulley et al.,
2016). A high magnitude event in the Deba catchment, Spain, however, led to a significant
redistribution of anthropogenically-enriched bed sediment metals (Martínez-Santos, Probst,
García-García & Ruiz-Romera, 2015). Further work is required to better understand changes
in sediment-associated metal concentrations and their spatial distributions in response to
flood events of various magnitudes.
Flood events may alter the sorption processes binding metals to sediment particles. Physical
resuspension of fine sediments may lead to desorption of associated contaminants (Zoumis,
Schmidt, Grigorova & Calmano, 2001) and flood events may also change water quality, such
as pH, which could alter sorption dynamics (Eggleton and Thomas, 2004). Such changes in
mobility influence the bioavailability of sediment-associated contaminants and this has
4
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
implications for ecosystems and the potential release of metals into the aquatic phase. The
dynamics of metal mobility during flood events and in flood-deposited overbank sediments
has been widely studied (Bradley, 1984; Leenaers, 1989b; Macklin and Dowsett, 1989;
Kozak, Skolasińska & Niedzielski, 2012). Typically, Fe and Mn oxides dominate in these
contexts (Macklin and Dowsett, 1989), particularly for Pb and Zn (Dawson and Macklin,
1998b). Flood events may also transform metals into more mobile forms (Leenaers, 1989b).
Only one study has examined bed sediment-associated metal partitioning following a flood
event and limited temporal variability was reported (Martínez-Santos et al., 2015). Further
work is required to explore the influence of extreme hydrological processes upon the mobility
of contaminated channel bed sediments. In light of projected increases in the frequency of
large flood events in response to climate change (Milly, Wetherald, Dunne & Delworth.,
2002; Prudhomme, Reynard & Crooks, 2002; Naylor et al., 2016), it is important to better
understand the mobilisation, export, and transformation of sediment-associated metal
concentrations associated with such flooding in a range of environmental contexts.
1.1. The Boxing Day 2015 flood event
During November and December 2015, a persistent low-pressure cyclonic system dominated
in northwest England. A number of large storm events took place in quick succession,
including Storm ‘Abigail’ (12th-13th November 2015), ‘Desmond’ (5th-6th December 2015),
and ‘Eva’ (24th December 2015). On the 26th December (Boxing Day) 2015, intense storm
conditions hit northern England. Within 36 hours 128 mm of rain fell (Holden Wood rain
gauge, Irwell catchment) – over 10% of the annual mean precipitation (1257 mm). These
storm events have been linked with a strongly positive phase of the North Atlantic Oscillation
and a strong El Niño bringing flooding to many parts of northern England and Scotland
5
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
(Marsh et al., 2016). Within the study catchments, these conditions led to widespread
saturated soils, which exacerbated flooding by increasing the flashiness of catchment
response (Parry et al., 2016). Overall, December was the wettest calendar month of 2015, and
2015 was the seventh wettest year on record in the UK (Met Office, 2016).
During the Boxing Day 2015 storm event, 37 out of 44 river level stations in the Irwell
catchment (793 km2) reached the highest levels on record. Figure 1 shows the exceedance
above historic peak river levels. The timing of peaks recorded across the Irwell showed early
maxima for the headwaters of the Roch (10:00-11:30) and the upper Irwell (12:00-13:00).
Stations further downstream peaked at 14:15-17:00, where some recorded an exceedance of
over 1 m above previous maxima. This includes the longest series at Adelphi Weir, which
dates back to 1949. Only seven of the 44 stations in the Irwell catchment did not record a new
peak river level, and six of these were very close to historical maxima. The instantaneous
peak discharge within the River Irwell was estimated to be nearly 700 m3 s-1 (Adelphi Weir;
17:00); however, flow measurements were beyond the existing rating curve at this site so
there is some uncertainty. Discharge may have in fact been as high as 900 m 3 s-1. Figure 2
shows the continuous flow record for the Adelphi Weir station (559.4 km2) in the lower
Irwell. The Boxing Day flood stands out as an exceptional event. Overall, 10 of the 12 flow
gauges in the Irwell catchment recorded new maximum discharges (GMCA, 2016). The peak
discharge from the River Irwell during the Boxing Day 2015 event is ranked in the top five
daily discharges on record in the UK (Barker, Hannaford, Muchan, Turner & Parry, 2016).
In contrast to the Irwell catchment, only two stations in the neighbouring Mersey catchment
(734 km2) registered record peak river levels during the Boxing Day flood (Figure 1).
Additionally, it took longer for high river levels to propagate through the upper Mersey
catchment. Peak discharge at Ashton Weir (660 km2), close to the outlet of the Mersey into
the Manchester Ship Canal, was 133 m3 s-1 (Figure 2). This was actually smaller than a
6
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
Figure 1: Exceedance of historic peak river level records during the Boxing Day 2015 flood event.
The Irwell and Mersey catchments are indicated by the red and green boundaries, respectively. The
timing of the peak river level recorded on Boxing Day 2015 are shown for each river level station.
The stations marked with a star correspond with the flow gauging stations in Figure 2. Data were
provided by the UK Environment Agency.
7
128
129
130
131
132
133
number of flood events in winter 2015/2016 (Figure 2). The geography of flooding across the
Manchester river network reflects the track of the storm and the pattern of upland soil
saturation in the headwaters and through the Irwell catchment. The Mersey catchment, with
headwaters located farther to the east, was less affected by the Boxing Day storm and
exhibited a greater lag in flood response. Thus, these neighbouring catchments experienced
markedly different hydrological conditions and flood magnitude during the Boxing Day 2015
event and winter 2015/2016 period. This provided a singular opportunity to examine the
impact of contrasting flood hydrology on fine-grained bed sediment storage and metal(loid)
dynamics in their respective channel networks.
Figure 2: Discharge data for a) Adelphi Weir in the Irwell catchment and b) Ashton Weir in the
Mersey catchment. The Boxing Day 2015 event is marked by a star. The locations of the gauging
8
134
135
136
137
138
139
140
141
142
143
144
145
stations are provided in Figure 1. The dates for pre- and post-flooding bed sediment sampling in each
catchment are indicated by the red bars. The grey shaded area (a) indicates a potential overestimation
of flow measurements, based upon the recording of discharge data that was beyond the rating curve
for this site. Data were provided by the UK Environment Agency.
1.2. Aims
This paper aims to:
1) examine the spatial redistribution of fine-grained bed sediment and associated
metal concentrations following the Boxing Day 2015 flood event and within a period of
sustained high flows in the Irwell and upper Mersey catchments; and
2) investigate the nature of any flood-related transformations in bed sediment metal
speciation.
2. Methods
2.1. Study area and sampling sites
The Irwell and upper Mersey catchments are underlain by Permo-Triassic sandstones and
mudstones in the southwest and coal measures in the north and west. Headwater streams to
the north and east drain the upland peatland environment of the southern Pennines. The rivers
range from an elevation of 462 m AOD in the Irwell headwaters to 10.3 m AOD where the
Mersey joins the Manchester Ship Canal. Mean annual precipitation is 1257 and 1150 mm for
the Irwell and upper Mersey catchments, respectively. These catchments are characterised by
meandering alluvial rivers with sandy gravel beds. Much of the river network passes through
9
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
urban and suburban land use. The history of the catchments is strongly tied to the Industrial
Revolution. Metal mining has never been important in these catchments although coal mining
has been locally important.
Channel bed sediments were sampled across both catchments, including headwater reaches.
Sampling encompassed 10 rivers: the Irwell, Roch, Tonge, Croal, Irk, Medlock, Mersey,
Tame, Etherow and Goyt (Figure 3). Fine-grained channel bed sediments were collected from
forty locations – these were selected to provide coverage that was representative of land use
types and local hydrogeomorphological settings.
2.2. Field sampling
Each site was sampled prior to and following the flooding of Winter 2015/16 (Figure 3).
Sampling was conducted under low flow conditions and during the same time of the year to
reduce the influence of any seasonal and hydrological variations on bed sediment storage and
sediment-associated contaminants. Pre-flood samples were collected between 18th April and
20th July 2015 and the same 40 sites were resampled between 7th May and 10th July 2016.
Fine-grained bed sediments were collected following the Lambert and Walling (1988)
method. A large cylinder (cross-sectional area: 0.14 m2) was eased 100 mm into the bed
sediment matrix. The bed sediment was then agitated to bring both surficial and interstitial
fine sediment into suspension using a trowel. The sediments were agitated for 10 s at 0, 2,
and 4 minutes at each site. The turbid water in the cylinder was then collected using plastic
jugs and transferred into a 25-l container. This procedure was performed at four locations
within each channel cross-section to provide a composite sample that was representative of
each site.
10
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
Figure 3: Location of the 40 bed sediment sampling sites. The Irwell and Mersey catchment
boundaries are shown in red and green, respectively. The 12 sites selected for sequential extraction
analysis are highlighted in the legend listing samples sites and on the map.
2.3. Sedimentological analyses
All sediments were wet sieved prior to analysis to isolate the <63 µm fraction and freeze-
dried. The preferential sorption capacity of silts and clays (<63 µm sediments) is well-
established in the context of contaminant studies (Förstner and Salomons, 1980; Horowitz,
1991). Grain size analysis was carried out using a Malvern 2000G Particle Sizer following
11
191
192
193
194
195
196
197
198
199
200
the removal of organic matter (by wet peroxide oxidation) and ultrasonic dispersion. Specific
surface area (SSA) was estimated by the Malvern software.
Following Owens, Walling & Leeks (1999), estimates of sediment storage of the <63 µm
fraction were produced using the equation:
BSs=C s W v
A
where bed sediment storage (BSs), reported as grams per square metre, is calculated as a
function of the sediment concentration associated with the container (Cs, g l−1) and the
volume of water enclosed in the cylinder (Wv, l), divided by the surface area of channel bed
that was isolated during the sampling procedure (A, 0.14 m2). Sediment storage was
calculated for pre- and post-flooding bed sediment samples to identify changes associated
with the flooding. This parameter represents the storage of fine-grained sediments in the
upper part of the bed sediment matrix that may be mobilised during a significant flood event.
2.4. Geochemical analyses
2.4.1. Total metal(loid) concentrations
This study builds upon an initial assessment of bed sediment quality in the Irwell and upper
Mersey catchments (Hurley et al., 2017) and we have focused on the same five
anthropogenically-enriched metal(loid)s: As, Cr, Cu, Pb, and Zn. Concentrations were
measured following an identical analytical procedure using XRF analysis with ICP-MS
calibration. XRF analysis permits a rapid assessment of metal(loid) concentrations and
provides an output encompassing a suite of analytes. Concentrations were obtained from
homogenised pressed powder briquettes and analysed using a Rigaku NEX-CG XRF.
12
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
Calibration was performed using standard reference materials (SRMs) NIST 2709 and 2780.
An additional calibration was applied to ensure comparability between XRF and ICP-MS
which was used for subsequent analyses. This included the analysis of a sub-set of samples
on ICP-MS which were used as an in-house library of standards in the post-processing
algorithm of the XRF quantification procedure. In effect, this supplies ‘known’
concentrations for the assessment of peak areas measured by the XRF analyser. All samples
subjected to sequential extractions (see below) were analysed for total metal(loid)s on ICP-
MS to ensure comparability and provide control checks on the efficiency of extractions.
For ICP-MS analysis, sediment samples (0.2 g) were digested in a Mars CEM microwave in
10 ml aqua regia (3:1 HCl to HNO3). Concentrations were measured using a PerkinElmer
NexION ICP-MS and a 7-point calibration procedure was applied. Certified SRMs were run
every 10 samples and drift was <10%. Calibration curves all had r2 values >0.999.
2.4.2. Total metal(loid) storage
The storage of each metal(loid) (Ms) on channel beds was estimated for each site as a
function of bed sediment storage (BSs; kg m−2) and metal(loid) concentration (Mc; mg kg−1)
following Walling, Owens, Foster, & Leeks, (2003). The value may be converted using the
dimensionless factor k to produce an output in g m-2, as is typical for other studies of metal
storage:
M S=BS s× M c
k
These results were compiled to produce a total metal(loid) channel bed storage value for both
the upper Mersey and Irwell catchments and the total combined catchment area. This was
calculated by extrapolating storage values from individual sampling sites to the reach-scale.
13
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
Reaches were defined as sections of the river network with similar hydrogeomorphological
properties. This was performed following Marttila and Kløve (2014) and Walling, Collins,
Jones, Leeks & Old (2006), who produced total catchment values for sediment storage. The
results were scaled up for each reach as a function of metal(loid) storage and channel area.
Channel area was established using Ordnance Survey MasterMap Water Network data and
field calculations applied in QGIS 2.18.0. Following the approach set out in Hurley,
Woodward & Rothwell (2018), flood-related change was assumed as the difference between
pre- and post-winter 2015/16 flooding values for the two catchments.
The uncertainty associated with catchment-scale estimates of total metal(loid) storage were
produced using the error from the metal(loid) analysis and that resulting from estimating
sediment storage, which was defined for the study rivers in Hurley et al. (2018).
2.4.3. Sequential extractions
A five-step modified Tessier, Campbell & Bisson (1979) sequential extraction procedure was
performed on a sub-set of the samples. This involved 12 bed sediment sample sites, analysed
for both pre- and post-flooding samples (n=24) (Figure 3). One gram of dried (<63 µm)
sediment was weighed into 50 ml nitric acid-washed, polyethylene tubes. Metal(loid)s were
then extracted in five operationally-defined phases:
1. Exchangeable fraction: 8 ml 1 M MgCl2
2. Acid soluble fraction (bound to carbonates): 8 ml 1 M NaOAc adjusted to pH 5.0 with
HOAc
3. Reducible fraction (bound to Fe and Mn oxides): 20 ml 0.4 M NH2OH.HCl in 25%
(v/v) HOAc
14
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
4. Oxidisable fraction (bound to organic matter): 3 ml 0.02 M HNO3 and 5 ml 30% H2O2
followed by a further 3 ml 30% H2O2 and finally extracted with 5 ml 3.2 M NH4OAc
in 20% (v/v) HNO3
5. Residual fraction: 10 ml aqua regia (3:1 HCl to HNO3)
Following extracts 1-4, samples were centrifuged for 25 min at 2000 rpm. The supernatant
was decanted for analysis whilst the residue was rinsed in deionised water for 15 minutes and
centrifuged for a further 15 to wash the sediments between stages. The residual fraction
(extract 5) was digested in a MARS CEM microwave for 1 hour before filtering and dilution.
All extracts were analysed using ICP-MS following the procedure detailed above. Duplicate
samples were subjected to the sequential extraction procedure and all results were within ±
10%.
To characterise concentration changes in sequential extraction data measured for pre- and
post-flooding bed sediment-associated metal(loid)s, a mobility factor (MF) was calculated
using the formula outlined by Kabala and Singh (2001):
MF= ( A+B )( A+B+C+ D+ E )
×100
where A, B, C, D, and E are the concentrations associated with the exchangeable, acid
soluble, reducible, oxidisable and residual fractions respectively. The MF represents the
proportion of the metal(loid) (A + B) that is considered to be more readily mobilised.
3. Results and discussion
3.1. Pre- and post-flooding sediment characteristics
15
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
Mean SSA and the <63 µm fraction (%) for the Irwell and upper Mersey catchments for pre-
and post-flooding datasets are shown in Figure 4. Grain size characteristics for the fine-
grained channel bed sediments show marked stability between pre- and post-flooding
datasets. No significant differences are observed in the % <63 µm fraction, for both the upper
Mersey (Mann Whitney-U: p = 0.522 sig level < 0.01) and Irwell (MW: p = 0.841 sig level <
0.01) catchments. There are no significant changes in the specific surface area (SSA) of the
<63 µm fraction of Irwell bed sediments following the flooding (MW: p = 0.136 sig level <
0.01). In the Mersey catchment, however, SSA demonstrates a significant increase (MW: p =
0.000 sig level < 0.01) indicating a fining of sediments within the silt and clay fraction. It is
notable that this difference is recorded within the upper Mersey catchment area that
experienced much more typical flood magnitudes within the sampling period.
In marked contrast to the grain size datasets, changes in the storage of fine-grained channel
bed sediments show considerable spatial variability pre- and post-flooding (Figure 4). Post-
flooding storage of silts and clays (<63 µm) varies ± 600% in comparison to pre-flooding
storage. There are no clear downstream trends in either catchments. These patterns
demonstrate that the period of sustained high flows in winter 2015/16, that included the
extreme Boxing Day flood event, had a significant impact on the spatial reorganisation of
fine sediments. In the Irwell catchment most sites show evidence of channel scour and a
significant fall in fine-grained sediment storage. Most channel bed sites in the upper Mersey,
where flooding was much less pronounced, show evidence of net fine sediment deposition.
Despite these striking changes in fine-grained sediment storage, the stability in the <63 µm
grain-size distribution at the majority of sites could indicate that a similar mix of fine-grained
sediment sources are activated during flood events of various magnitudes. Sources of
sediment to the river channel include channel erosion, hillslope soil erosion, and urban
runoff. The upper Mersey and Irwell catchments are meandering alluvial rivers, where a large
16
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
Figure 4: Changes in channel bed sediment storage (<63 µm) following the flood event. Proportional
circles denote the % difference between pre- and post-flood values. Averaged grain size parameters
for each catchment are also provided for the <63 µm size fraction.
proportion of fine-grained sediment is likely to be derived from bank erosion. The catchment-
wide contrasts in fine-grained sediment storage may relate to different hydrological
conditions – perhaps influenced by direct river channel modifications including channel
straightening and artificial bank protection in urban and suburban reaches. The lack of clear
downstream trends suggests that local sources of sediment may continue to be important,
even in exceptional flood events.
3.2. Total bed sediment metal(loid) concentrations
Figure 5 shows the pre- and post-flooding concentrations for the selected metal(loid)s.
Concentrations demonstrate markedly conservative behaviour, where ‘conservative’ is
defined here as exhibiting negligible change. Mann Whitney-U tests reveal no significant
17
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
Figure 5: Total metal(loid) concentrations for channel bed sediments prior to and following the
Boxing Day 2015 flood event. Note that scales for the proportional circles vary between the selected
metal(loid)s. The darker segment represents pre-flood values and the lighter colour post-flood values.
18
330
331
332
333
differences between pre- and post-flooding concentrations for each of the metal(loid)s: As: p
= 0.689; Cr: p = 0.303; Cu: p = 0.149; Pb: p = 0.936; Zn: p = 0.401 (sig level < 0.01).
Moreover, there are no significant differences within the upper Mersey and Irwell catchments
and the spatial patterning remains remarkably similar across the entire study area. The
distribution of pre-flooding bed sediment-associated metal(loid)s and their potential sources
are discussed in Hurley, Rothwell, & Woodward (2017). Hurley et al. (2017) also compared
pre-flooding values to available sediment quality guidelines (SQGs). The post-flooding
concentrations do not alter the original assessment, where all sites exceed threshold and
probable effects levels (TELs and PELs) for all of the selected metal(loids) for both Canadian
and draft UK SQGs (CCME, 2002; Hudson-Edwards, Macklin, & Brewer, et al., 2008). This
highlights the limited influence of the flooding upon any potential risks posed by sediment-
associated metal(loid) contamination in the Mersey and Irwell system.
In common with the pre-flooding samples, the post-flooding sediment characteristics
demonstrate no obvious controls on metal concentrations. For example, there is no correlation
between the selected metal(loid)s and grain size (Pearson’s: p > 0.1). This suggests that the
pattern of metal(loid) sources across the study area has not been significantly altered during
the flood event. A small number of sites, however, do exhibit more complex behaviour in
their response to the flooding. Sites which exhibit changes for one metal(loid) may show very
little difference in others. These include, for example, increases in As and Cr around Bolton,
an increase in Zn concentrations along the Irk, and increases for all metal(loid)s at some key
contaminated sites such as Bradshaw, Denton, and Irwell Springs (Figure 5). The Bradshaw
site (Figure 3; Site 14) in the Croal catchment presents the greatest change, with post-flood
As and Cr concentrations 3 times higher than pre-flooding values and a twofold increase in
Cu concentration. There are a number of notable increases in the upper Mersey, including the
highest Zn values (representing a 2.6x increase) observed at the Denton site (Figure 3; Site
19
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
30) on the River Tame, despite the more typical flooding that occurred in that catchment. The
absence of clear downstream patterns suggests these concentrations are the product of highly
localised contamination sources. Further study is needed to better understand the processes
responsible for these localised contamination increases and their limited propagation
downstream.
Table 1 highlights the extent of the conservative behaviour of metal(loid)s within the Irwell
and upper Mersey catchments. Over 50% of post-flooding metal(loid) concentrations are
within ± 20% of pre-flooding values and approximately one third of sample sites exhibit
changes of less than ± 10%. These include reaches in both the Mersey and Irwell catchments,
with no observable influence associated with the extreme nature of the flooding in the Irwell.
This suggests a degree of stability in the sources delivering contaminants to the fluvial
environment. Moreover, it is likely that these sources represent significant long-term inputs
of metal(loid)s, which are not exhausted during a high magnitude flooding. To gain a better
understanding of the dominant sediment sources in these catchments the sediment
fingerprinting approach could be employed (Walling and Woodward, 1995; Owens et al.
2016). The limited change in contamination pattern across the catchments indicates that the
operation of these sources is not discharge-dependent. The Irwell contains a greater number
of sites with post-flood concentrations lying within ± 10% and ± 20% of the pre-flooding
values. In other words, the catchment that experienced the biggest flood shows the most
conservative behaviour.
Published studies report a wide variety of channel bed contamination responses to flood
events. Several studies record significant differences in pre- and post-flood bed sediment
metals (Protasowicki, Niedzagwiecki, Ciereszko, Perkowska, & Meller, 1999; Moody,
Sullivan, & Taylor, 2000; Ciszewski, 2001; Symader and Roth, 2002; Martínez-Santos et al.,
2015; Wadige et al., 2016). Symader, Bierl, & Hampe (1994) report greater changes
20
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
Table 1: The proportion of sites where post-flooding total metal(loid) concentrations are within ± 10%
and ± 20% of pre-flooding values. This is shown by catchment and for the total study area.
Irwell Mersey Irwell & Mersey±10% ±20% ±10% ±20% ±10% ±20%
As 42 % 58 % 29 % 43 % 38 % 53 %Cr 35 % 73 % 29 % 71 % 33 % 73 %Cu 31 % 62 % 36 % 64 % 33 % 63 %Pb 31 % 69 % 43 % 57 % 38 % 65 %Zn 35 % 58 % 36 % 57 % 35 % 58 %
associated with winter flood events, which has not been observed in the Boxing Day 2015
event. The results reported here do, however, corroborate data presented for other urbanised
catchments. For example, Horowitz et al. (2014) recorded no significant change in channel
bed sediment contamination in 54 rivers of the US Atlantic coast following Hurricane Irene
and Tropical Storm Lee (both 2011). Despite significant volumes of sediment being
discharged from these rivers, post-flooding bed sediment-associated metals strongly reflected
pre-flooding spatial patterns. This also indicates a degree of stability in catchment sediment
sources and, in particular, the resuspension of channel bed sediments or local overbank
stores, where river engineering may have limited the downstream distribution of freshly
eroded sediments (Horowitz et al., 2014). Additionally, studies by Pulley et al. (2016) and
Hutchinson and Rothwell (2008) both demonstrate stability in suspended sediment
concentrations associated with historical or urban contamination with persistent metal sources
delivering material to the channel during a range of hydrological conditions.
Given the succession of high magnitude storm events that occurred in the north of England
during November–December 2015, these data suggest that the dominant sources of
contaminants across the catchments are remarkably stable. The very wet antecedent
conditions did not appear to deplete metal stores, a feature since no substantial decreases in
metal concentration are evident across all 40 bed sediment sampling sites. This may also
21
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
reflect the replenishment of urban-derived contamination between runoff events (Pulley et
al., 2016) in addition to the reworking of spatially extensive stores of historically-
contaminated floodplain sediments. Carter et al. (2003) show that channel banks can provide
an important source of sediment within urbanised river catchments. Whilst river engineering
is prominent across the Mersey-Irwell system, this is not on the scale evident in many US
coastal rivers. A potentially important mechanism contributing to this limited response might
be the significant contamination of headwater systems during the industrial period. Much of
the early industry in Manchester was concentrated in upstream reaches to utilise stream flows
for water power and capture the ‘cleaner’ waters for processes such as textile bleaching and
dyeing (Hurley et al., 2017). Disposal of physical and chemical waste directly into river
channels during the 19th and early 20th century is likely to have produced a network of
contaminated floodplain environments in headwater catchments. Moreover, atmospheric
deposition associated with coal-powered industry has contaminated headwater soils and
peatlands with a range of metal(loid)s, including Pb and As (Rothwell et al., 2005, 2009).
This very probably limits the availability of ‘clean’ sediments to dilute contaminated loads
downstream. The Boxing Day flood event may have reworked a good deal of this
historically-contaminated material, therefore leading to negligible changes in bed sediment-
associated metals across the catchments after the winter of 2015/16.
3.3. Bed sediment-associated storage of metal(loid)s
Estimates for the mean sediment-associated storage of metal(loid)s across the two study
catchments are given in Table 2. Note that these data refer to the storage of metal(loid)s
within the <63 µm fraction on the channel bed. These values are similar to those observed for
22
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
Table 2: Mean sediment-associated storage of metal(loid)s (g m -2) on channel beds in the Irwell (a)
and Mersey (b) catchments and the total study area (c). Ranges are provided in brackets.
a. Irwell catchment
b. Mersey catchment
c. Irwell and Mersey catchments combined
Pre-flooding Post-floodingAs 0.004 (0.000-0.023) 0.004 (0.000-0.028)Cr 0.021 (0.003-0.137) 0.023 (0.002-0.130)Cu 0.023 (0.002-0.292) 0.020 (0.001-0.076)Pb 0.036 (0.004-0.269) 0.031 (0.004-0.107)Zn 0.087 (0.011-0.642) 0.090 (0.009-0.431)
the Swale, Aire and Calder catchments in northern England: 0.0004-0.4532 g m -2 (Cr);
0.0013-0.3403 g m-2 (Cu); 0.018-0.497 g m-2 (Pb); 0.07-1.31 g m-2 (Zn) (Walling et al., 2003).
Sediment-associated storage of As has not been reported by any previous study. Studies that
have investigated metal(loid) storage associated with the sediment phase have observed that
the amount of metal(loid) storage is more strongly controlled by sediment storage than by
metal(loid) concentrations, although concentrations are still important (Walling et al., 2003;
Collins, Walling, & Leeks, 2005; Estrany, Garcia, Walling, & Ferrer, 2011).
23
Pre-flooding Post-floodingAs 0.003 (0.001-0.009) 0.002 (0.000-0.006)Cr 0.015(0.006-0.064) 0.015 (0.002-0.047)Cu 0.015 (0.004-0.050) 0.011 (0.001-0.036)Pb 0.033 (0.008-0.103) 0.024 (0.004-0.076)Zn 0.081 (0.019-0.287) 0.060 (0.009-0.157)
Pre-flooding Post-floodingAs 0.005 (0.000-0.023) 0.006 (0.001-0.028)Cr 0.023 (0.003-0.137) 0.027 (0.005-0.130)Cu 0.027 (0.002-0.292) 0.025 (0.004-0.076)Pb 0.038 (0.004-0.269) 0.035 (0.006-0.107)Zn 0.090 (0.011-0.642) 0.106 (0.016-0.431)
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
The mean metal(loid) storage associated with fine sediments across the catchments does not
vary significantly between pre- and post-flooding samples: As: p = 0.834; Cr: p = 0.638; Cu:
p = 0.617; Pb: p = 0.928; Zn: p = 0.726 (Mann Whitney-U; sig level <0.01). It has been noted
elsewhere that contaminant storage is likely to vary temporally in line with seasonal sediment
dynamics and metal(loid) supply to the channel environment (Walling et al., 2003; Collins et
al., 2005). Our samples were collected at the same time of year in 2015 and 2016, to
eliminate seasonal variability from the results. Despite the marked variability in <63 µm
sediment storage across individual sites between the pre- and post-flooding samples,
metal(loid) storage does not differ significantly when averaged across the catchments.
Figure 6 presents the change in total metal(loid) storage associated with the flooding. A net
decrease is observed for the total study area, although the decrease is relatively small (0.29-
14.2%) when compared to pre-flooding catchment-wide storage. When broken down by
catchment, the difference between pre- and post-flooding values indicates that between the
two sampling periods, the Mersey was characterised by net accumulation and the Irwell by
net loss. This may relate to the different hydrological conditions within the two catchments,
where the Irwell experienced much higher magnitude flooding. Note, however, that the
uncertainty estimates indicate that the absolute difference between pre- and post-flooding
values is negligible. This concurs with the sediment-associated metal(loid) concentration
data, which point towards the conservative nature of metal(loid)s following the Boxing Day
2015 flooding.
24
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
Figure 6: Storage of the selected metal(loids) for the Mersey and Irwell catchments for pre- and post-
flooding sampling (a) and change in channel bed storage of the metal(loid)s between pre- and post-
flooding values across the entire river network (b). Error bars are based on the calculated uncertainty
associated with sediment storage estimation and metal(loid) analysis
3.4. Sequential extraction of metal(loid) concentrations
Sequential extractions of total metal(loid)s were performed on 12 sites for both pre- and post-
flooding sediments. Figure 7 compares the pre- and post-flooding sequential extraction
results. Hurley et al. (2017) discuss the relative proportions of concentrations across the
operationally-defined fractions. There is very limited variability within the metals across the
12 sites where Fe and Mn oxides are the dominant scavengers of Pb and Zn, organic matter is
important for Cu, and both the reducible and residual fractions are important for Cr and As.
25
469
470
471
472
473
474
475
476
477
478
479
480
481
Figure 7: Pre- and post-flood sequential extraction results for 12 selected channel bed sediment sites
in the Irwell and Mersey catchments. The mobility factors for each site and metal(loid) are provided
for both pre- and post-flooding data. Site numbers refer to those listed in Figure 3.
26
482
483
484
485
Figure 7 also presents the mobility factors for pre- and post-flooding metal(loid) extractions.
Small increases in mobility are observed for Cr, Zn, and Pb, whilst As and Cu are more
variable. Changes in mobility are associated with the exchangeable and acid soluble fractions.
However, across the Mersey and Irwell catchments, the exchangeable fraction actually
decreases across all of the selected metal(loid)s. Instead, the MF is primarily controlled by
increases in the acid soluble fraction, where this is associated with decreases in the relative
proportion of the reducible fraction for As and Zn, the oxidisable fraction for Cu, and the
residual fraction for Cr and Pb. It is notable that the mobility factor changes are relatively
small and that the most mobile fraction, the chemically exchangeable fraction, is actually
reduced. This suggests there are no major long-lasting effects related to the release of metals
from bed sediments in response to major flooding.
It is interesting to note that despite the magnitude of the flooding observed within the Irwell
catchment, transformations to the metal(loid)s stored within channel bed sediments are
minimal and largely indiscernible from those observed in the Mersey catchment, which
experienced a more typical period of flooding. This suggests that large flood events either do
not strongly influence the chemical partitioning of metal(loid)s bound to sediments or that
these systems quickly recover from disturbance events. A comparable outcome was reported
by Martínez-Santos et al. (2015) who showed that changes in the chemical partitioning of
metals following a high magnitude flood event in the Deba catchment, northern Spain, were
negligible. Temporal variability was associated with increases in the residual fraction, which
may have been linked to reductions in the organic matter content of sediments.
The marked consistency in metal partitioning further indicates stability in the sources of
metal contaminated sediments. This is exemplified by the Denton site (Figure 3: site 30)
where the sequential extraction data show only very minimal changes despite a substantial
increase in total metal concentrations in the post-flood dataset. These results suggest that the
27
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
spatial network of contaminant stores is a more important control on fine-grained bed
sediment contaminant patterns than hydrological processes. Moreover, the evident stability in
contaminant sources across successive flood events indicates that the observed channel bed
contamination is likely to persist into the future since these sources show no sign of
depletion. This has implications for river quality improvement efforts in these reaches, as
well as for downstream environments receiving fluvial sediment inputs.
4. Conclusions
Despite a dramatic reorganisation of channel bed sediment storage in the Irwell and upper
Mersey fluvial systems, sediment-associated metal(loid) concentrations display markedly
conservative behaviour following the high magnitude Boxing Day 2015 flood event and a
period of sustained high flows in winter 2015/16. There is no significant difference between
pre- and post-flooding metal(loid) concentrations. Furthermore, there is no discernible
difference in the response observed in the Irwell compared to that in the upper Mersey
catchment, where the latter experienced a much more typical period of flooding. Given the
dramatic improvements in water quality in these catchments since the 1980s (Burton, 2003),
the enduring sediment contamination issue points towards sources that are ‘locked in’, and
potentially reinforced by the distinctive hydrological and geomorphological conditions of
urban river networks. These contaminant sources may include the reworking of historically-
contaminated alluvial materials which could help to explain persistent elevated
concentrations, even in headwater catchments. This suggests that poor sediment quality may
continue to be a feature of the Irwell and upper Mersey catchments for many decades.
28
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
6. Acknowledgments
The authors wish to thank all those who assisted with the fieldwork. We would also like to
thank John Moore, Jonathan Yarwood, and Thomas Bishop in the Geography Laboratories
for assistance with a range of analyses. We thank the two anonymous reviewers for their
constructive reviews of our paper. River flow data were provided by the Environment
Agency.
5. References
Baborowski M., & Einax J. W. 2016. Flood-event based metal distribution patterns in water
as approach for source apportionment of pollution on catchment scale: Examples from
the River Elbe. Journal of Hydrology, 535, pp.429–437.
Barker L. J., Hannaford J., Muchan K. G. L., Turner S. P., & Parry S. 2016. The winter
2015/2016 floods in the UK: a hydrological appraisal. Weather, 71 (12), pp.324–333
Bradley S. B. 1984. Flood effects on the transport of heavy metals. International Journal of
Environmental Studies, 22 (3–4), pp.225–230.
Bradley S. B., & Lewin J. 1982. Transport of heavy metals on suspended sediments under
high flow conditions in a mineralised region of wales. Environmental Pollution Series
B, Chemical and Physical, 4 (4), pp.257–267.
Burton, L. R. 2003. The Mersey Basin: an historical assessment of water quality from an
anecdotal perspective. Science of the Total Environment, 214-316, pp.53-66.
Carter, J., Owens, P. N., Walling, D. E., & Leeks, G. J. L. 2003. Fingerprinting suspended
sediment sources in a large urban river system. Science of the Total Environment,
314-316, pp.513-534.
29
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
CCME. 2002. Canadian sediment quality guidelines for the protection of aquatic life.
Available at: http://www.ecy.wa.gov/programs/eap/psamp/BoundaryBay/PSAMP-
BBAMP%20documents/Canadian%20guidelines%20for%20water%20quality/
SedimentProtAquaticLifeSummaryTables(en).pdf. [Accessed 20 Sept 2017].
Ciszewski D. 2001. Flood-related changes in heavy metal concentrations within sediments of
the Bial̶a Przemsza River. Geomorphology, 40 (3–4), pp.205–218.
Collins, A. L., Walling, D. E, & Leeks, G. J. L. 2005. Storage of fine-grained sediment and
associated contaminants within the channels of lowland permeable catchments in the
UK. In Sediment budgets: proceedings of the International Symposium on Sediment
Budgets (Proceedings of the Seventh Scientific Assembly of the International
Association of Hydrological Sciences (IAHS) at Foz do Iguaço, Brazil, 3-9 April,
2005), IAHS Press, 291, pp.269-291.
Coynel A., Schäfer J., Blanc G., & Bossy C. 2007. Scenario of particulate trace metal and
metalloid transport during a major flood event inferred from transient geochemical
signals. Applied Geochemistry, 22 (4), pp.821–836.
Dawson E. J., & Macklin M. G. 1998a. Speciation of heavy metals on suspended sediment
under high flow conditions in the River Aire, West Yorkshire, UK. Hydrological
Processes, 12 (9), pp.1483–1494.
Dawson E. J., & Macklin M. G. 1998b. Speciation of heavy metals in floodplain and flood
sediments: a reconnaissance survey of the Aire Valley, West Yorkshire, Great Britain.
Environmental Geochemistry and Health, 20 (2), pp.67–76.
Eggleton J., & Thomas K. V. 2004. A review of factors affecting the release and
bioavailability of contaminants during sediment disturbance events. Environment
International, 30 (7), pp.973–980.
30
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
Estrany, J., Garcia, C., Walling, D. E., & Ferrer, L. 2011. Fluxes and storage of fine-grained
sediment and associated contaminants in the Na Borges River (Mallorca, Spain).
CATENA, 87 (3), pp.291-305.
Förstner U., & Salomons W. 1980. Trace metal analysis on polluted sediments.
Environmental Technology Letters, 1 (11), pp.494–505.
Hjulström, F. 1935. Studies of the morphological activity of rivers as illustrated
by the River Fyris. Geological Institute Upsala, 25, pp.221-527.
Horowitz A. J. 1991. A primer on sediment-trace element chemistry. Lewis Publishers.
Horowitz A. J., Elrick K. A., & Smith J. J. 2008. Monitoring urban impacts on suspended
sediment, trace element, and nutrient fluxes within the City of Atlanta, Georgia, USA:
program design, methodological considerations, and initial results. Hydrological
Processes, 22 (10), pp.1473–1496.
Horowitz A. J., Elrick K. A., Smith J. J., & Stephens V. C. 2014. The effects of Hurricane
Irene and Tropical Storm Lee on the bed sediment geochemistry of U.S. Atlantic
coastal rivers. Hydrological Processes, 28 (3), pp.1250–1259.
Hudson-Edwards, K. A., Macklin, M. G., Brewer, P. A., & Dennis, I. A. 2008. Assessment of
metal mining-contaminated river sediments in England and Wales. Environment
Agency, Bristol, UK
Hurley R. R., Woodward J. C., & Rothwell J. J. 2018. Microplastic contamination of river
beds significantly reduced by catchment-wide flooding. Nature Geoscience, 11 (4),
pp.251-257.
Hurley R. R., Rothwell J. J., & Woodward J. C. 2017. Metal contamination of bed sediments
in the Irwell and Upper Mersey catchments, northwest England: exploring the legacy
of industry and urban growth. Journal of Soils and Sediments 17 (11), pp.2648-2665.
31
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
Hutchinson S. M., & Rothwell J. J. 2008. Mobilisation of sediment-associated metals from
historical Pb working sites on the River Sheaf, Sheffield, UK. Environmental
Pollution, 155 (1), pp.61–71.
GMCA (2016). Flood Investigation Report: Greater Manchester 26th December 2015.
Available at:
https://www.greatermanchester-ca.gov.uk/downloads/file/199/boxing_day_flood_rep
ort_2015 [Accessed 27th September 2016].
Kabala C., & Singh B. R. 2001. Fractionation and Mobility of Copper, Lead, and Zinc in Soil
Profiles in the Vicinity of a Copper Smelter. Journal of Environment Quality, 30 (2),
pp.485-492.
Kozak L., Skolasińska K., & Niedzielski P. 2012. Environmental impact of flood: the study
of arsenic speciation in exchangeable fraction of flood deposits of Warta river
(Poland) in determination of “finger prints” of the pollutants origin and the ways of
the migration. Chemosphere, 89 (3), pp.257–261.
Lambert C. P., & Walling D. E. 1988. Measurement of channel storage of suspended
sediment in a gravel-bed river. CATENA, 15 (1), pp.65–80.
Lee J. H., Bang K. W., Ketchum Jr. L. H., Choe J. S, & Yu M. J. 2002. First flush analysis of
urban storm runoff. Science of The Total Environment 293 (1–3), pp.163–175.
Leenaers H. 1989a. The transport of heavy metals during flood events in the polluted river
Geul (the Netherlands). Hydrological Processes 3 (4), pp.325–338.
Leenaers H. 1989b. Downstream changes of total and partitioned metal concentrations in the
flood deposits of the river Geul (The Netherlands). GeoJournal 19 (1): pp.37–43.
Macklin M. G., & Dowsett R. B. 1989. The chemical and physical speciation of trace metals
in fine grained overbank flood sediments in the Tyne basin, north-east England.
CATENA 16 (2), pp.135–151.
32
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
Marsh T. J., Kirby C., Muchan K. G. L., Barker L. J., Henderson E., & Hannaford J. 2016.
The winter floods of 2015/2016 in the UK - a review Available at:
http://nrfa.ceh.ac.uk/occasional-reports [Accessed 8 June 2017].
Marrtila, H. & Kløve, B. 2014. Storage, properties and seasonal variations in fine‐grained bed
sediment within the main channel and headwaters of the River Sanginjoki, Finland.
Hydrological Processes 28 (17), pp. 4756-4765.
Martínez-Santos M., Probst A., García-García J., & Ruiz-Romera E. 2015. Influence of
anthropogenic inputs and a high-magnitude flood event on metal contamination
pattern in surface bottom sediments from the Deba River urban catchment. Science of
The Total Environment 514, pp.10–25.
Met Office. Annual 2016. Met Office Available at:
http://www.metoffice.gov.uk/climate/uk/summaries/2015/annual [Accessed 1
December 2016]
Miller J., Barr R., Grow D., Lechler P., Richardson D., Waltman K., & Warwick J. 1999.
Effects of the 1997 flood on the transport and storage of sediment and mercury within
the Carson River Valley, West‐Central Nevada. The Journal of Geology 107 (3):
pp.313–327.
Milly P. C. D., Wetherald R. T., Dunne K. A., & Delworth T. L. 2002. Increasing risk of
great floods in a changing climate. Nature 415 (6871): pp.514–517.
Moody J. A., Sullivan J. F., & Taylor H. E. 2000. Effects of the flood of 1993 on the
chemical characteristics of bed sediments in the Upper Mississippi River. Water, Air,
and Soil Pollution 117 (1–4): pp.329–351.
Naylor L. A., Spencer T., Lane S. N., Darby S. E., Magilligan F. J., Macklin M. G., & Möller
I. 2016. Stormy geomorphology: geomorphic contributions in an age of climate
extremes. Earth Surface Processes and Landforms 42 (1): pp.166–190.
33
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
Old G. H., Leeks G. J. L., Packman J. C., Smith B. P. G., Lewis S., & Hewitt E. J. 2006.
River flow and associated transport of sediments and solutes through a highly
urbanised catchment, Bradford, West Yorkshire. Science of The Total Environment
360 (1–3): pp.98–108.
Owens P. N., Walling D. E., & Leeks G. J. L. 1999. Deposition and storage of fine-grained
sediment within the main channel system of the River Tweed, Scotland. Earth Surface
Processes and Landforms 24 (12): pp.1061–1076.
Owens, P. N., Blake, W. H., Gaspar, L., Gateuille, D., Koiter, A. J., Lobb, D. A., Petticrew,
E. L., … Woodward, J. 2016. Fingerprinting and tracing the sources of soils and
sediments: Earth and ocean science, geoarchaeological, forensic, and human health
applications. Earth-Science Reviews 162: pp.1-23,
Parry S., Barker L., Prosdocimi I., Lewis M., Hannaford J., & Clemas S. 2016. Hydrological
summary for the United Kingdom: December 2015 Available at:
http://nora.nerc.ac.uk/512654 [Accessed 18 April 2017]
Peraza-Castro M., Sauvage S., Sánchez-Pérez J. M., & Ruiz-Romera E. 2016. Effect of flood
events on transport of suspended sediments, organic matter and particulate metals in a
forest watershed in the Basque Country (Northern Spain). Science of The Total
Environment 569–570: pp.784–797.
Protasowicki M., Niedzagwiecki E., Ciereszko W., Perkowska A., & Meller E. 1999. The
comparison of sediment contamination in the area of estuary and the lower course of
the Odra before and after the flood of summer 1997. Acta hydrochimica et
hydrobiologica 27 (5): pp.338–342.
Prudhomme C., Reynard N., & Crooks S. 2002. Downscaling of global climate models for
flood frequency analysis: where are we now? Hydrological Processes 16 (6):
pp.1137–1150.
34
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
Pulley S., Foster I., & Antunes P. 2016. The dynamics of sediment-associated contaminants
over a transition from drought to multiple flood events in a lowland UK catchment.
Hydrological Processes 30 (5): pp.704–719.
Resongles E., Casiot C., Freydier R., Le Gall M., & Elbaz-Poulichet F. 2015. Variation of
dissolved and particulate metal(loid) (As, Cd, Pb, Sb, Tl, Zn) concentrations under
varying discharge during a Mediterranean flood in a former mining watershed, the
Gardon River (France). Journal of Geochemical Exploration 158: pp.132–142.
Rothwell J. J., Robinson S. G., Evans M. G., Yang J., & Allott T. E. H. 2005. Heavy metal
release by peat erosion in the Peak District, southern Pennines, UK. Hydrological
Processes, 19 (15), pp.2973–2989.
Rothwell J. J., Taylor K. G., Ander E. L., Evans M. G., Daniels S. M., & Allott T. E. H.
2009. Arsenic retention and release in ombrotrophic peatlands. Science of The Total
Environment, 407 (4), pp.1405–1417.
Symader W., & Roth M. 2002. Changes in chemical characteristics of river bed samples
caused by exceptional high floods in the Kartelbornsbach basin near Trier.
International Association of Hydrological Sciences, Publication, 276, pp.333–338
Symader W., Bierl R., & Hampe K. 1994. Temporal variations of organic micropollutants
during storm events in a small river catchment. In Hydrological, Chemical, and
Biological Processes of Transformation and Transport of Contaminants in Aquatic
Environments (Proceedings of the Rostov-on-Don Symposium, May 1993), IAHS
Press,, 219, pp.423–428
Tessier A., Campbell P. G. C., & Bisson M. 1979. Sequential extraction procedure for the
speciation of particulate trace metals. Analytical Chemistry. 51 (7), pp.844–851.
35
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
Wadige C. P. M. M., Taylor A. M., Krikowa F., & Maher W. A. 2016. Sediment metal
concentration survey along the mine-affected Molonglo River, NSW, Australia.
Archives of Environmental Contamination and Toxicology, 70 (3), pp.572–582.
Walling, D. E., Collins, A. L., Jones, P. A., Leeks, G. J. L., & Old, G. 2006. Establishing
fine-grained sediment budgets for the Pang and Lambourne LOCAR catchments, UK.
Journal of Hydrology, 330 (1-2), pp.126-141.
Walling D. E., Owens P. N., Carter, J., Leeks, G. J. L., Lewis, S., Meharg, A. A., & Wright,
J.(2003). Storage of sediment-associated nutrients and contaminants in river channel
and floodplain systems. Applied Geochemistry, 18 (2), pp.195-220.
Walling D. E., & Woodward J. C. 1992. Use of radiometric fingerprints to derive information
on suspended sediment sources. Erosion and sediment transport monitoring
programmes in river basins, 210, pp.153–164
Walling, D. E. & Woodward, J. C. 1995. Tracing sources of suspended sediment in river
basins: a case study of the River Culm, Devon, UK. Marine and Freshwater Research
46 (1), 327-336.
Walling D. E., Webb B. W., & Woodward J. C. 1992. Some sampling considerations in the
design of effective strategies for monitoring sediment-associated transport. Erosion
and sediment transport monitoring programmes in river basins, 210, pp.279–288.
Yin C., & Li L. 2008. An investigation on suspended solids sources in urban stormwater
runoff using 7Be and 210Pb as tracers. Water Science & Technology, 57 (12), pp.1945-
1950.
Zonta R., Collavini F., Zaggia L., & Zuliani A. 2005. The effect of floods on the transport of
suspended sediments and contaminants: A case study from the estuary of the Dese
River (Venice Lagoon, Italy). Environment International, 31 (7), pp.948–958.
36
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
Zoumis T., Schmidt A., Grigorova L., & Calmano W. 2001. Contaminants in sediments:
remobilisation and demobilisation. Science of The Total Environment, 266 (1–3),
pp.195–202.
37
727
728
729
730