Environmental Pollution 214 (2016) 314e323

Contents lists available at ScienceDirect

Environmental Pollution
journal homepage: www.elsevier.com/locate/envpol

In-situ characterization and assessment of arsenic mobility in lake

sediments*
Qin Sun a, Shiming Ding b, *, Yan Wang b, Lv Xu a, Dan Wang b, Jing Chen a,
Chaosheng Zhang c
a
Key Laboratory of Integrated Regulation and Resource Development on Shallow Lakes, Ministry of Education, College of Environment, Hohai University,
Nanjing 210098, China
b
State Key Laboratory of Lake Science and Environment, Nanjing Institute of Geography and Limnology, Chinese Academy of Sciences, Nanjing 210008,
China
c
GIS Centre, Ryan Institute and School of Geography and Archaeology, National University of Ireland, Galway, Ireland

a r t i c l e i n f o a b s t r a c t

Article history: In-situ characterization and assessment of arsenic (As) mobility in sediments was scarce. In this study,
Received 10 January 2016 the distributions of labile As at a vertical resolution of 2 mm were obtained in the sediments of a large
Received in revised form Lake Taihu through in-situ measurements using a Zr-oxide diffusive gradients in thin lms (Zr-oxide
10 April 2016
DGT) technique. The DGT-labile As, interpreted as DGT ux (FDGT), exhibited three different patterns in
Accepted 11 April 2016
the lake, with all the patterns generally showing an increasing mobility followed by a decreasing
mobility with sediment depth. The mobility of As could be characterized by the average FDGT (0.06
e1.27 pg cm
2 s
1) in the top 10 mm surface sediments, the maximal FDGT (FDGT-M, 0.14
Keywords:
Arsenic
e2.44 pg cm
2 s
1) in the end of the initial increasing phase of FDGT, and the diffusion length (DL, 28
Sediment e66 mm) from the depth showing the FDGT-M to the sediment-water interface. The upward mobilization
Mobility of labile As from the deep sediments to the surface sediments and overlying water became evident when
Zr-oxide DGT FDGT-M > 1.7 pg cm
2 s
1 or DL < 41 mm. The results, for the rst time, showed a prospect in in-situ risk
Eutrophication assessment of the pollution of sediment As. It was suggested that the increasing mobility of As in the
High resolution upper sediments was controlled by the reduction of As(V) and the reductive dissolution of Fe(III) (hydr)
oxides, while the decreasing mobility in the deep sediments was attributed to immobilization of As(III)
by secondary Fe(II)-bearing minerals.
2016 Elsevier Ltd. All rights reserved.

1. Introduction
Arsenic (As) is a well-known pollutant receiving great attention
because of its prevalence in the environment and its toxicity to
organisms. Besides natural pollution from geogenic sources, spe-
cic attention has been paid to anthropogenic contribution of As in
certain areas (Fendorf et al., 2010; Moriarty et al., 2014). The major
anthropogenic sources of As include mining, smelting of non-
ferrous metals and burning of fossil fuels, use of arsenic-
containing pesticides, and use of arsenic in the preservation of
timber (Smedley and Kinniburgh, 2002). Freshwaters are at risk of
As contamination due to external input of As from these
*
This paper has been recommended for acceptance by Eddy Y. Zeng.
* Corresponding author.
E-mail address: smding@niglas.ac.cn (S. Ding).
anthropogenic activities (Jay et al., 2005), which are likely inten-
sied in eutrophic lakes due to the potential release of As from
sediments (Rahman and Hasegawa, 2012). According to earlier in-
vestigations by Sohrin et al. (1997) and Hasegawa et al. (2010) on 18
eutrophic lakes in Japan, concentrations of As in summer with
persisting anoxia were 2e7 times higher than those in winter.
Barringer et al. (2011) presented a similar nding on a eutrophic
Lake Mohawk, in which concentrations of As in summer were
higher than those in winter by a magnitude. Hasegawa (1997) and
Sohrin et al. (1997) observed simultaneous increases of dissolved
Fe, Mn and As in the water column, based on which the authors
attributed the increase of water As to the release of As from sedi-
ments under anoxic conditions. Since many eutrophic lakes are
used for drinking waters, it calls on a health concern and the ne-
cessity in monitoring and assessment of the mobility of As in
sediments.
Until now, characterization of As mobility in sediments is mostly

http://dx.doi.org/10.1016/j.envpol.2016.04.039
0269-7491/ 2016 Elsevier Ltd. All rights reserved.
 Q. Sun et al. / Environmental Pollution 214 (2016) 314e323
performed through ex situ experiments, typically including
collection of sediment samples followed by chemical extraction of
operationally dened fractions of As. However, it is hard to avoid
exposing sediments to the air during the operation process, which
has a risk of changing As speciation through oxygenation of As(III)
and Fe(II) (Wilkie and Hering, 1998). The chemical extraction
method has also been long questioned because of poor precision
and selectivity resulting from operational inconsistency and species
redistribution during extraction and separation (Wang and
Mulligan, 2008). The method does not consider the dynamic
character of As during its releases from the sediment solids to pore
water, which is a necessary step prior to the mobilization of As to
the overlying water (Hossain et al., 2012). Furthermore, the sedi-
ment has great heterogeneity in chemical distributions especially
in the vicinity of the sediment-water interface (SWI) (Stockdale
et al., 2009), but the traditional method has a relatively low
spatial resolution (cm) in measurement. To our knowledge, there
are few reports on in-situ assessment of As mobility in sediments
by avoiding the above shortages.
Diffusive gradients in thin lms (DGT) is a passive sampling
technique capable of tracking labile As fractions from pore water
and the resupply from sediment solids at a high spatial resolution
(Zhang et al., 2014). Its measurement relies on a diffusion process of
inorganic As ions from the surface of the DGT device to a binding
phase through a diffusion layer (Sun et al., 2014). The DGT has been
developed to characterize the distribution of total inorganic As, to
investigate As speciation and to explore the mechanisms of FeeAs
interaction in freshwater, estuarine and marine sediments (Bennett
et al., 2012a; Gao et al., 2009; Zhang et al., 2002). Until now, the
uses of DGT has been mostly limited to sediment cores collected
from the eld, while there are a few on-site applications (Ding et al.,
2015; Wang et al., 2016).
Taihu is the 3rd largest lake in China. It is a typical eutrophic
lake, with the water trophic level declining from the North and
Northwest to South and Southeast. Meanwhile, it has regions
dominated by algal blooms and macrophytes mainly in the North
and Southeast bays (Qin et al., 2007). The status of As in Lake Taihu
has been investigated recently by Wei et al. (2011). The total con-
centration of As in the water column in spring, 2009 ranged from 15
to 70 mg L
1, which surpassed the WHO limit (10 mg L
1). The total
content of As in sediments varied from 30 to 75 mg kg
1 and
exceeded the geochemical baseline (9.92 mg kg
1) of the sediments
in Lake Taihu. It seems that contamination of As has became a
problem in this lake, especially under the background that Taihu
functions as an important source of drinking water. Consequently, it
is necessary to carry out an in-situ study on the mobility of As in
sediments in this lake, since the release of As from the sediment can
signicantly affect the level of As in the water column (Jay et al.,
2005; Sohrin et al., 1997; Hasegawa et al., 2010; Barringer et al.,
2011).
The present study carried out an in-situ investigation into the
distribution and mobility of labile As in the sediments of Lake Taihu
using an established Zr-oxide DGT (Sun et al., 2014). The DGT
probes were deployed at a total of 14 sites to obtain the distribution
of labile As in sediments at a 2 mm spatial resolution. The upward
mobility of labile As was quantitatively assessed based on the
extracted parameters from the measurements. The possible
mechanisms involved in the distribution and mobility of labile As
were discussed.
2. Materials and methods
2.1. Description of the sampling sites
In this study, a total of 14 sites were selected representative of
315
the status of the lake (Fig. 1) (Ding et al., 2015). Sites 1 to 5 were
located in the north and northwest regions frequently dominated
with algae. Sites 10 to 14 were located in the southeast bays
dominated with macrophytes. The other sites had no visible
macrophyte coverage but occasionally suffered from algal blooms.
The variation of ecological type and water quality status in Lake
Taihu were reported elsewhere (Duan et al., 2009; Ma et al., 2008).
2.2. Field sampling
The principle of DGT together with modication of the Zr-oxide
DGT is explained in Supplementary material. The Zr-oxide binding
gel was provided by Easysensor Ltd. (www.easysensor.net), which
was prepared according to Ding et al. (2013). The gel was covered
with a cellulose nitrate membrane (0.45 mm pore size, 0.11 mm
thickness, Millipore), which was used as a thin diffusion layer. They
were sealed in a holder made by Perspex. A sheet of sponge was
attached on the back of the holder to make it easier to mark the
sediment-water interface (SWI) after retrieving the probe from the
sediments (Ding et al., 2015). The Zr-oxide DGT probes were
deoxygenated with nitrogen overnight. They were stored in a
container lled with deoxygenated NaCl solution (0.1 M) and
transported to each sampling site through speedboats. The probes
were inserted in the sediments using a releasing device in
October 2014 and deployed for 24 h (Fig. 1). After retrieval, each
probe was rinsed by lake water and then deionized water to remove
visible sediment particles attached. They were placed in a container
sealed at air temperature to prevent moisture loss and transported
to the laboratory.
The overlying water within a distance of 20 cm to the SWI was
collected in each site using a self-made sampler. The sediment
cores were also collected using a gravity sampler. The surface
sediments with a thickness of 50 mm were sectioned, which were
mainly formed from the sedimentation during the last 20 years
assuming an average sedimentation rate of 2.1 mm yr
1 in Taihu
(Qin et al., 2007). The water and sediment samples were stored in
~10  C and transported to the laboratory. Temperature in water was
simultaneously measured twice in each site at the deployment and
retrieval of the DGT probes, respectively.
2.3. Sample analysis
The Zr-oxide gels were disassembled from the deployed DGT
probes. Each binding gel was sliced at a 2-mm interval using a
cutter made by stacking ceramic blades with a thickness of 1.0 mm.
Each slice was eluted using 400 mL of 1.0 M NaOH (Sun et al., 2014).
Each elution solution was added with 5% KI plus 5% ascorbic acid
(w/v, prepared freshly) at pH 5 buffered by 0.25 M HAc-NaAc and
left at room temperature for at least 30 min to reduce As(V) to
mez-Ariza et al., 2000). Concentra-
As(III) (Correia et al., 2010; Go
tion of total As was then determined using online, intermittent
ow, hydride generation coupled with AFS (HG-AFS) (AF-610D,
Beijing Rayleigh Analytical Instrument Corporation, China) through
formation of arsine between As(III) and KBH4 (Sun et al., 2014).
The sediments were frozen at
20  C and freeze-dried
under
80  C. Dried sediments were ground and passed through
a 2-mm sieve. A portion of each sediment sample was digested
with aqua regia for the detection of total As using HG-AFS. The
accuracy of the analysis was checked with an error of 8.2% using
standard reference material for lake sediments (GBW07436, Center
for Standard Reference of China). Another portion of the sediment
was used for fractionation of As speciation according to Wang et al.
(2010), which divides As in sediments into non-specically sorbed
(F1), specically sorbed (F2), Al oxide/oxyhydroxides associated
(F3), amorphous and poorly crystalline Fe oxide/oxyhydroxides
316 Q. Sun et al. / Environmental Pollution 214 (2016) 314e323

Fig. 1. Sampling in Lake Taihu, including the distribution of sampling sites (left) and a photograph of the deployment of a DGT probe using a releasing device in the eld (right). Sites
1 to 5 are frequently dominated with alga, while Sites 10 to 14 are macrophyte-dominated. Other sites had no visible macrophyte coverage but occasionally suffered from algal
blooms.

bound (F4), well crystalline Fe oxide/oxyhydroxides bound (F5) and
residual forms (F6). Contents of amorphous and crystalline Fe in
sediments were simultaneously detected in F4 and F5 using an ICP-
AES.
Total As in unltered and ltered (through a 0.45 mm pore size
membrane) water samples was determined using HG-AFS. The
unltered water was digested using concentrated HNO3 prior to As
determination. The accuracy of the analysis was checked using
standard reference material for natural waters (GBW 080390,
Center for Standard Reference of China). The average deviation was
less than 3% through analyses of 8 parallels.
Organic matter in the sediments was determined using the
potassium dichromate-dilution heat colorimetric method (Bao and
Jiang, 1999). The pH was analyzed in a 1:10 solid:liquid ratio sus-
pension using a pH electrode.
2.4. Data proceeding
The DGT-labile As was interpreted as a ux (FDGT, pg cm
2 s
1)
(Santner et al., 2015):
M sites of No. 1e6, demonstrating a pollution character in these sites
FDGT 1000
t due to industrial and domestic wastewater drainage (Qin et al.,
The t (s) is the deployment time of DGT. M is the accumulation
mass of As in the binding gel (M, ng cm
2), which is calculated
using eq. (2):


Ce Ve Vg
M (2)
Afe
where Ce is the concentration of As in the elution solution
(ng ml
1), Ge is the elution efciency (0.877) (Sun et al., 2014), A is
the surface area of each gel slice (cm2), and Ve and Vg are the vol-
umes of elution solution and gel slice (ml), respectively.
3. Results and discussion
3.1. General properties of the sediments and water
The general properties of the sediments and overlying water are
shown in Table 1. The pH values of the sediments varied in a range
of 6.64e7.07, showing a weakly acidic to neutral character. The
amount of OM showed a strong variation ranging from 0.35% to
1.96%. High values were observed at two sites with coverage of
rooted submerged plants (Sites 13 and 14), followed by those sites
in the polluted northern bays frequently dominated by algal
blooms (Sites 1e3). It is evident that high productivity caused
accumulation of OM in the sediments of these sites. Contents of
amorphous and crystalline Fe in sediments ranged from 8.48 to
16.02 mg kg
1 and 626.5e2266.1 mg kg
1, respectively. High con-
tents appeared at Sites 1 to 3 in the northern bays. All the sediment
samples contained high content of silt (73.1e84.2%), followed by
clay (14.6e26.0%) and sand (0.4e5.8%). A signicant correlation
was found between sand and crystalline Fe (r 0.789, p 0.001). It
veried that Fe preferred to be enriched in ne-grained clay
(Bishop et al., 2014).
The concentration of dissolved oxygen in the overlying water
ranged from 7.44 to 10.32 mg L
1, reecting an oxic condition in
benthic bottom. The water pH values were weak alkaline ranging
between 7.94 and 8.73. The concentrations of total N and total P
varied from 0.534 to 2.719 mg L
1 and from 0.015 to 0.222 mg L
1,
respectively. High concentrations were observed in the northern
(1)
2007).
3.2. Total arsenic in sediments and water
The concentrations of total dissolved As and total As in the
overlying water were in the ranges of 0.45e4.29 mg L
1 and
6.23e15.9 mg L
1, respectively (Table 2). The concentrations of total
As in nearly a half of the sites exceeded the WHO limit (10 mg L
1),
but they were under the permissible limit (50 mg L
1) for the rst
class water source dened by Environmental Quality Standards for
Surface Water (GB 3838-2002) in China. High concentrations
appeared at the northern sites of 1e4, suggesting a pollution
character as observed based on results of total N and total P
(Table 1). The concentration range of total dissolved As in the water
was similar to 1.4e5.7 mg L
1 detected in May 2010 by Zhang et al.
(2013), and was also at a low level in comparison with Lake Dianchi
(3.1e10.5 mg L
1) and many other lakes summarized by Zhang et al.
(2013).
 Q. Sun et al. / Environmental Pollution 214 (2016) 314e323 317

Table 1
Basic properties of the surface sediments and overlying water of Lake Taihu.

NO. Sediment Overlying water

pH OM Amorphous Fe Crystalline Fe Particle size distribution DO pH TN TP

1
1
1
1
(%) (mg kg ) (mg kg ) <4 mm 4e64 mm 64e1024 mm (mg L ) (mg L ) (mg L
1)

1 6.84 1.40 14.77 2266.1 20.6 78.2 1.2 8.86 7.94 1.938 0.219
2 7.07 0.90 13.27 1233.6 20.8 75.2 4.0 9.03 8.43 2.331 0.222
3 6.97 0.86 14.39 1619.3 26.0 73.1 0.8 7.44 8.08 2.719 0.191
4 6.96 0.40 16.02 1237.5 19.9 77.7 2.4 9.83 8.43 1.863 0.175
5 6.88 0.58 9.08 1128.3 21.1 78.5 0.4 9.82 8.73 1.018 0.095
6 6.73 0.71 13.43 1096.5 16.8 79.2 4.1 9.24 8.35 1.987 0.163
7 6.98 0.63 12.52 977.9 16.3 81.9 1.8 9.31 8.21 0.681 0.038
8 6.85 0.54 14.05 993.1 15.7 79.4 4.9 9.87 8.26 0.573 0.023
9 6.75 0.35 10.12 1021.5 17.9 81.7 0.4 9.37 8.38 0.975 0.064
10 6.70 0.51 8.48 626.5 16.2 78.0 5.8 9.39 8.27 0.534 0.033
11 6.99 0.37 14.87 984.1 17.2 78.6 4.2 9.66 8.25 0.591 0.015
12 7.04 0.68 13.09 1128.7 20.3 77.8 1.8 9.43 8.33 0.632 0.048
13 6.91 1.25 10.44 708.7 14.6 84.2 1.2 10.05 8.33 0.579 0.052
14 6.64 1.96 11.4 1122.9 15.7 79.5 4.8 10.03 8.22 0.780 0.058
Ave. 6.9 0.8 13.3 1153.2 18.5 78.8 2.7 9.42 8.30 1.240 0.101
Min. 6.64 0.35 8.48 626.5 14.6 73.1 0.4 7.44 7.94 0.534 0.015
Max. 7.07 1.96 16.02 2266.1 26 84.2 5.8 10.3 8.73 2.719 0.222

Table 2
Concentrations of total As and As fractions in sediments as well as total As and total dissolved As in the overlying water.

NO. Sediment Overlying water

Total (mg kg
1) As speciation (mg kg
1) Dissolved (mg L
1) Total (mg L
1)
a
F1 F2 F3 F4 F5 F6

1 14.5 0.32 3.12 1.60 5.62 3.79 1.16 4.29 14.8

2 13.6 0.47 2.02 1.32 5.97 1.44 1.68 1.10 15.5
3 17.3 0.42 2.80 1.15 6.68 2.43 2.72 2.61 14.8
4 10.1 0.17 1.71 0.86 6.38 1.23 0.95 4.20 15.9
5 10.6 0.18 1.66 1.25 4.01 1.42 1.58 3.09 11.4
6 9.60 0.18 1.51 0.77 4.32 1.10 0.52 4.06 10.1
7 13.5 0.29 1.80 0.77 9.65 1.29 1.21 1.10 7.68
8 11.5 0.03 2.14 1.07 4.19 0.83 1.33 2.12 8.35
9 11.8 0.17 2.06 3.05 5.46 1.07 1.18 2.03 8.60
10 6.42 0.16 1.65 1.00 2.17 0.33 1.09 0.61 6.23
11 9.80 0.21 1.51 1.13 3.36 2.01 0.98 2.07 7.70
12 9.17 0.14 1.17 1.22 4.62 0.63 0.67 1.54 7.34
13 10.7 0.32 3.05 2.18 3.46 0.39 0.81 1.76 10.1
14 6.81 0.21 1.02 0.75 2.92 0.73 0.88 0.45 8.13
Ave. 11.1 0.23 1.94 1.29 4.92 1.33 1.20 2.22 10.5
Min. 6.42 0.03 1.02 0.75 2.17 0.33 0.52 0.45 6.23
Max. 17.3 0.47 3.12 3.05 9.65 3.79 2.72 4.29 15.9
a
Arsenic in sediments were divided into non-specically sorbed (F1), specically sorbed (F2), Al oxide/oxyhydroxides associated (F3), amorphous and poorly crystalline Fe
oxide/oxyhydroxides bound (F4), well crystalline Fe oxide/oxyhydroxides bound (F5) and residual (F6) forms.

The content of total As in the sediments ranged from 6.42 to
17.3 mg kg
1, with an average of 11.1 mg kg
1. The total As in the
sediments were positively correlated with the contents of amor-
phous Fe (r 0.673, p 0.008). Such a correlation has been
observed in the surface sediments of the Pearl River in Southern
China and a ooded soil in Rocky Mountain arsenal area (Wang
et al., 2010; Corwin et al., 1999), demonstrating that Fe cycling
may play a dominant role in the partition of As in the sediments.
The levels of As detected in the sediments of Lake Taihu reached
4.66e10.85 mg kg
1 reported by Zhang et al. (2013), but they were
lower than 7.4e64 mg kg
1 reported by Qu et al. (2001). This
should be attributed to the removal of contaminated sediments
through dredging in the northern bays of Lake Taihu after 2000 (He
et al., 2013).
3.3. Arsenic speciation in sediments
Arsenic in the sediments was sequentially fractionated into 6
chemical forms (F1 to F6), with the concentrations listed in Table 2.
They were in the ranges of 0.03e0.47, 1.02e3.12, 0.75e3.05,
2.17e9.65, 0.33e3.79 and 0.52e2.72 mg kg
1, with mean values of
0.23, 1.94, 1.29, 4.92, 1.33, 1.20 mg kg
1, respectively. They
accounted for 0.3e3.7%, 12.0e29.9%, 5.1e23.5%, 33.9e64.3%,
3.8e24.3% and 6.2e16.8% of total As in the sediments, respectively
(Fig. S1). Their average contributions were ranked in the order of
F4 > F2 > F3 z F5 z F6 > F1. The sum of the fractions F4 (amor-
phous and poorly crystalline Fe bound form) and F5 (well crystal-
line Fe bound form) accounted for 56.4% of the total As in the
sediments on average. It revealed that As was primarily associated
with Fe (hydr)oxides in the sediments, which is consistent with the
correlation between the contents of total As and amorphous Fe as
observed earlier.
3.4. Distribution patterns of DGT-labile As
It should be noted that the As measured by the Zr-oxide DGT is
the total inorganic As (Sun et al., 2014). Inorganic As has been re-
ported as the predominant form of As in pore waters of freshwater
318 Q. Sun et al. / Environmental Pollution 214 (2016) 314e323

sediments, while organic As species are often undetectable or
contributing to 10% of the total arsenic measured (Huerga et al.,
2005; Liao and Deng, 2006; He et al., 2007; Bennett et al., 2012b).
The distributions of DGT-labile As, interpreted as FDGT, are listed in
Fig. 2. Generally, the vertical distribution of FDGT in the sediments
can be classied into 3 patterns for the 14 sites investigated, with
the schematic diagram showing in Fig. 3. In the rst pattern
(Pattern I, the same below), the FDGT remained steady or at small
uctuation to a depth of
10 to
20 mm (the minus symbol rep-
resents the depth below the SWI) and increased sharply to a middle

Fig. 2. Distribution of DGT-labile As (interpreted as FDGT) in the sediment-overlying water proles of Lake Taihu. The dotted lines at the depth of zero show the position of the
sediment-water interface.
 Q. Sun et al. / Environmental Pollution 214 (2016) 314e323 319

depth followed by a decrease to the bottom of the sediments. A
total of 9 sites (Sites 4e12) belonged to this pattern. Among these
sites, 6 sites were located in the regions without serious algal
blooms, while 3 sites were covered by macrophytes (mainly
oating plants). In Pattern II, the distribution of FDGT was similar to
that of Pattern I, but with a steady stage or small uctuation of FDGT
following the downward increase of FDGT. Two sites (13 and 14)
belonged to this pattern, which were located in the region covered
by submerged macrophytes. In Pattern III, an increase of FDGT
appeared from the SWI to the middle depth, followed by a steady
stage and then a decrease to the bottom of the sediments. Sites 2
and 3 showed this pattern. An exception was found for Site 1, in
which the FDGT did not decrease following the stable stage of FDGT.
Such a change was regarded as a variant of the Pattern III, since the
FDGT likely decreased in deeper sediment layer. The three sites were
located in the northern bays with pollution.
Previous studies have reported the DGT-labile As distributions
in the sediments of Scottish sea-loch, Belgian coast, Australian Gold
Coast and Lake Hongze in China, which can be mostly grouped to
Pattern I and then Pattern III (Bennett et al., 2012a; Gao et al., 2009;
Zhang et al., 2002). The distribution patterns of labile As observed
in this study should be thus representative of other aquatic envi-
ronments. Different from the measurements using sediment cores
collected from the eld (Bennett et al., 2012a; Gao et al., 2009;
Zhang et al., 2002), this study provided in-situ observation of the
As distribution patterns in sediments.
from the depth of the FDGT-M to the SWI was dened as DL (Fig. 3). A
greater FDGT-M with a shorter DL reected a greater diffusion po-
tential of As from the deep sediments to the surface sediments and
thus a greater pollution risk to the overlying water. For the 9 sites
belonging to Pattern I, the FDGT-M varied from 0.14 to
1.77 pg cm
2 s
1, with an average of 0.58 pg cm
2 s
1. The corre-
sponding DL ranged from 32 to 66 mm, with an average of 51 mm
(Table 3). For the remaining 5 sites belonging to Pattern II and III,
the FDGT-M varied from 1.33 to 2.44 pg cm
2 s
1. Their average was
0.95 pg cm
2 s
1, which was much greater than that of Pattern I.
The corresponding DL ranged from 28 to 56 mm. Their average
value (34 mm) was much shorter than that of Pattern I. A signi-
cantly negative correlation was further found between FDGT-M and
DL for the 13 sites investigated (Fig. S2). It demonstrated that the
sites of Pattern II and III were at higher risk of As pollution in the
overlying water in comparison with the sites of Pattern I, due to a
greater potential of As ux from the deep sediments.
3.6. Upward mobilization of DGT-labile As
The DGT-measured FDGT at each depth reects the localized
lability of As in sediments including pore water concentration and

3.5. Key features of the DGT-labile As proles

The key features of the DGT-labile As proles were summarized

in Table 3, including the distribution patterns of FDGT as mentioned
above. In the overlying water at the depth of 0 (i.e., the SWI) to
10 mm, the average FDGT values varied from 0.07 to
0.48 pg cm
2 s
1, with greater values (0.36e0.48 pg cm
2 s
1) at
Sites 1e3 in the northern bays. In the surface sediment layers at the
depth of 0 to
10 mm, the average FDGT values varied from 0.06 to
1.27 pg cm
2 s
1. Greater values (0.42e1.27 pg cm
2 s
1) appeared
at Sites 1e3 and 7 located in the northern region.
In the deep sediment layer at the depth below
10 mm, all the
three distribution patterns of FDGT exhibited an increasing stage of Fig. 3. Schematic illustration of the distribution and mobility of labile As in sediments.
FDGT with sediment depth. Here, the maximal FDGT at the end of the The proles I, II and III represent the distribution patterns of labile As in uncontami-
increasing stage was dened as FDGT-M, while the diffusion length nated, submerged macrophyte-covered and contaminated sites, respectively.

Table 3
The key features extracted from the DGT proles for characterization of As mobility in sediments.

NO. Distribution pattern Overlying water (0e10 mm depth) Surface sediment (0 to
10 mm depth) Deep sediment
(<
10 mm depth)
a a
I II III Average FDGT Average FDGT Parameter
b
FDGT-M c
DL
1 C 0.48 0.65 1.82 40
2 C 0.39 0.86 2.44 32
3 C 0.36 1.27 2.29 28
4 C 0.12 0.12 0.21 48
5 C 0.15 0.14 0.18 60
6 C 0.13 0.12 0.18 56
7 C 0.21 0.42 1.77 56
8 C 0.19 0.19 0.71 40
9 C 0.16 0.15 0.75 42
10 C 0.09 0.09 0.53 62
11 C 0.09 0.08 0.74 66
12 C 0.07 0.06 0.14 32
13 C 0.19 0.18 1.61 38
14 C 0.10 0.12 1.33 34
a
Average values of FDGT in the dened depths, with the unit of pg cm
2 s
1.
b
The maximal values of FDGT at the end of increasing stage in the FDGT distribution patterns.
c
The diffusion length (mm) from the depth appearing the FDGT to the SWI.
320 Q. Sun et al. / Environmental Pollution 214 (2016) 314e323
the dynamic resupply from sediment solids. Since an increasing
phase of FDGT appeared in the three distribution patterns with
sediment depth, there could be an upward mobilization of labile As
from deep sediment layer. In order to reveal the mobilization
process, the relationships among the FDGT in the sediment and
overwater layers were investigated (Fig. 4).
It is evident that the FDGT in the surface sediments remained
steady at values below 0.2 pg cm
2 s
1 with the increase of the
FDGT-M in deep sediments (Fig. 4A). When the FDGT-M was greater
than 1.7 pg cm
2 s
1 (or DL < 41 mm calculated using equation in
Fig. S2), the FDGT in the surface sediments had a sharp increase until
1.3 pg cm
2 s
1. A similar trend was observed for the FDGT in the
overlying water with the change of FDGT in the surface sediments
(Fig. 4B). When the FDGT in the surface sediments was below
0.2 pg cm
2 s
1, a linear correlation (p < 0.001) was observed with
the FDGT in the overlying water. The slope of 1.03 in the tting
equation reected that there was an exchange equilibrium of labile
As between in the two layers, which is consistent with the non-
gradient distributions of labile As in the vicinity of the SWI in the
major sites (Fig. 2). When the FDGT in the surface sediments was
greater than 0.2 pg cm
2 s
1, the increase of FDGT in the surface
sediments became stronger than that in the overlying water. This
change originated from the sharp increase of FDGT in the surface
sediments when the FDGT-M was greater than 1.7 pg cm
2 s
1
(Fig. 4A).
Consequently, a high FDGT-M in the deep sediments drove an
upward mobilization of labile As, leading to the increase of FDGT in
the surface sediments and then in the overlying water. It reected a
visible pollution of labile As from the deep sediments when it was
accumulated to a threshold mass in deep sediments. This was
supported by the upward ux gradients of labile As in the surface
sediments as well as in the overlying water as observed in the sites
of Pattern III (Sites 1e3) (Fig. 2).
3.7. Relationships between DGT-labile As and overlying water As
The relationships of the DGT-labile As (FDGT) in the deep sedi-
ments, surface sediments and overlying water with the concen-
trations of total dissolved As and total As in the overlying water are
shown in Fig. 5. No correlations were found for the total dissolved
As with the FDGT in the three compartments, which may attribute to
the low concentration and small variation of total dissolved As in
the overlying water (Table 2). Positive correlations were observed
Fig. 4. Relationships of the average DGT ux (FDGT) in the surface sediments (below
10 mm depth) with the maximal FDGT (FDGT-M) in deep sediments (A) and the average FDGT in
the overlying water (0e10 mm depth) (B), respectively. The vertically dot line shows the FDGT-M at 1.7 pg cm
2 s
1. The horizontally dot line shows the FDGT in the surface sediments
at 0.2 pg cm
2 s
1. The three asterisks show the signicant level of p < 0.001.
for the total As, with the signicant level increased in the order of
deep sediments, surface sediments and overlying water. It dem-
onstrates that the DGT-labile As was a sensitive indicator in
reection of the impact of sediment As on overlying water As. This
has recently been veried in Lake Hongze measured by Zr-oxide
DGT (Wang et al., 2016). The increase in the signicant level
along the upward direction reected a decreasing impact of labile
As from the deep sediments, with the impact increased signi-
cantly when FDGT-M > 1.7 pg cm
2 s
1 or DL < 41 mm as observed
earlier (Fig. 4).
3.8. Possible mechanisms of DGT-labile As distribution and
mobilization
Three general patterns were observed in the proles of labile As
(Figs. 2 and 3). All of them had a common feature, reected by a
downward increasing of FDGT with sediment depth. Such a feature
has been observed with DGT measurements in marine sediment
cores collected from Scottish sea-loch and Belgian coast (Gao et al.,
2009; Zhang et al., 2002), in two sediment cores collected from a
freshwater river and an estuarine on the Gold Coast, Australia
(Bennett et al., 2012a), and in situ DGT measurement in Lake
Hongze, China (Wang et al., 2016). A simultaneous increase of
dissolved Fe(II) in porewater was observed by Bennett et al.
(2012a), reected by a positive correlation between DGT-labile As
and porewater Fe(II). The increase of labile As with sediment depth
is generally attributed to reductive dissolution of As-bearing Fe(III)
(hydr)oxides with the transition of the sedimentary redox potential
status from oxic to anoxic (Bose and Sharma, 2002; Campbell et al.,
2008; Pedersen et al., 2006). This is consistent with an earlier
nding that As primarily associated with Fe (hydr)oxides in sedi-
ments (Fig. S1).
The northern sites (Sites 1e3) with pollution character con-
tained relatively high contents of organic matter and total As in the
sediments and low concentrations of dissolved oxygen in the
overlying water (Tables 1 and 2), which facilitated the reduction of
As(V) and Fe(III) and resulted in higher FDGT in top surface sedi-
ments and greater FDGT-M with shorter DL in deep sediments
(Table 3). The favorable conditions might cause a persistent
reduction of Fe(III) (hydr)oxides in deep sediments, leading to the
formation of the steady phase of FDGT following its increasing phase
as observed in Pattern III. The steady phase of high FDGT represented
a large source of labile As for diffusion upward to the overlying
 Q. Sun et al. / Environmental Pollution 214 (2016) 314e323 321

Fig. 5. Relationships of measured DGT ux (FDGT) in deep sediments (FDGT-M, at the depth below
10 mm), surface sediments (average FDGT, at the depth of 0 to
10 mm) and
overlying water (average FDGT, at the depth of 0e10 mm) with concentrations of total dissolved As and total As in overlying water, respectively. The data from Site 4 were excluded
from statistic analysis due to its anomaly relative to other data; the one and two asterisks show the signicant levels at p < 0.05 and 0.01, respectively.

water, resulting in the strong diffusion gradients of FDGT in the top
surface sediments on Site 1e3 (Fig. 2). The two submerged
macrophytes-covered sites (Sites 13, 14) also contained relatively
high contents of organic matter in sediments, causing high and
persistent FDGT in deep sediments as observed (Fig. 2). However, the
FDGT in the top surface sediments and overlying water remained at a
low level, which should be attributed to codeposition with organic
carbon (Meharg et al., 2006). Furthermore, macrophyte roots could
secrete oxygen to create localized oxic environment and immobi-
lize As onto Fe(III) (hydr)oxides (Hupfer and Dollan, 2003; Williams
et al., 2014).
A decreasing phase of FDGT was observed in all the 3 distribution
patterns. Similar phenomena have been reported in other sediment
cores (Bennett et al., 2012a; Gao et al., 2009; Zhang et al., 2002;
Campbell et al., 2008), but the mechanisms remained to be iden-
tied. It has been hypothesized that the decrease of FDGT was due to
322 Q. Sun et al. / Environmental Pollution 214 (2016) 314e323
immobilization of inorganic As(III) by Fe(II)- and Fe(II)/Fe(III)-
bearing minerals (e.g., magnetite, siderite, and green rust) form-
ing concomitantly during microbial reduction of Fe(III) (Campbell
et al., 2008; Handley et al., 2013; Cummings et al., 2000; Root
et al., 2007). Handley et al. (2013) studied the changes of As
solid-phase associations during Fe redox transformations in a
sediment with a low As:Fe ratio (1:1000 wt/wt). The authors found
that 40% of inorganic As initially bound by poorly-crystalline iron
(hydr)oxides in sediments had been increasingly incorporated into
well-crystallized Fe(II)-bearing minerals or specically-sorbed to
the mineral surfaces with sediment depth and decreasing Eh
values. The Fe(II)-bearing minerals had been identied primarily as
ferrous iron carbonates (e.g., siderite), while natural siderite had
been recognized as an efcient mineral for arsenic retention in
sediments (Guo et al., 2007). In addition, Root et al. (2007) found
that As was immobilized in an iron-rich sediment from a drinking-
water reservoir accompanied by the formation of reduced mixed-
valency green rust. In Lake Taihu, the average content of Fe in
sediments was reported at 32.4 mg g
1 (Bai et al., 2009). The
average As:Fe ratio was calculated at 1:2919 wt/wt using an average
As content of 11.1 mg kg
1 detected in this study. This ratio is lower
than that of the sediments of Handley et al. (2013). It implies that
the immobilization process of As(III) by Fe(II)- and Fe(II)/Fe(III)-
bearing minerals should be stronger in deep sediments.
4. Conclusions
The mobility of As in sediments and its potential contamination
on water column As were successfully characterized and assessed
based on the use of the in situ, high-resolution DGT technique. The
results show that the content of total As in the sediments of Lake
Taihu ranged from 6.42 to 17.3 mg kg
1, with on average 54% of the
total As associated with Fe (hydr)oxides. The DGT-labile As
exhibited three basic patterns, with all the patterns generally
showing an increasing mobility followed by a decreasing mobility
with sediment depth. A visible pollution of labile As appeared on
the water column from the deep sediments when FDGT-

2
1
M > 1.7 pg cm s or DL < 41 mm. Positive correlations were
further found for the total As in the overlying water with the
average FDGT in the surface and deep sediments, demonstrating that
the DGT-labile As is a sensitive indicator in reection of the impact
of sediment As on overlying water As.
Acknowledgments
The authors thank Prof. Weiping Hu for providing speed boats in
eld sampling and Dr. Paul N. Williams for improving the quality of
this manuscript. This study was jointly sponsored by the National
Scientic Foundation of China (41322011, 51479068, 41571465), a
fund from the Priority Academic Program Development of Jiangsu
Higher Education Institutions (PAPD), and the State Key Laboratory
of Lake Science and Environment (2014SKL011).
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.envpol.2016.04.039.
References
Bai, X.L., Ding, S.M., Fan, C.X., Liu, T., Shi, D., Zhang, L., 2009. Organic phosphorus
species in surface sediments of a large, shallow, eutrophic lake, Lake Taihu,
China. Environ. Pollut. 157 (8e9), 2507e2513.
Bao, S.D., Jiang, R.F., 1999. Measurement of Organic Matter Content in Soil. Analysis
of Agri-chemistry in Soil, third ed. Chinese Agricultural Science and Technology
Press, Beijing, pp. 34e35.
Barringer, J.L., Szabo, Z., Wilson, T.P., Bonin, J.L., Kratzer, T., Cenno, K., Romagna, T.,
Alebus, M., Hirst, B., 2011. Distribution and seasonal dynamics of arsenic in a
shallow lake in northwestern New Jersey, USA. Environ. Geochem. Health 33
(1), 1e22.
Bennett, W.W., Teasdale, P.R., Panther, J.G., Welsh, D.T., Zhao, H.J., Jolley, D.F., 2012a.
Investigating arsenic speciation and mobilization in sediments with DGT and
DET: a mesocosm evaluation of oxic-anoxic transitions. Environ. Sci. Technol. 46
(7), 3981e3989.
Bennett, W.W., Teasdale, P.R., Welsh, D.T., Panther, J.G., Stewart, R.R., Price, H.L.,
Jolley, D.F., 2012b. Inorganic arsenic and iron(II) distributions in sediment
porewaters investigated by a combined DGT-colourimetric DET technique. En-
viron. Chem. 9 (1), 31e40.
Bishop, M.E., Glasser, P., Dong, H.L., Arey, B., Kovarik, L., 2014. Reduction and
immobilization of hexavalent chromium by microbially reduced Fe-bearing clay
minerals. Geochim. Cosmochim. Acta 133, 186e203.
Bose, P., Sharma, A., 2002. Role of iron in controlling speciation and mobilization of
arsenic in subsurface environment. Water Res. 36 (19), 4916e4926.
Campbell, K.M., Root, R., O'Day, P.A., Hering, J.G., 2008. A gel probe equilibrium
sampler for measuring arsenic porewater proles and sorption gradients in
sediments: II. Field application to Haiwee Reservoir sediment. Environ. Sci.
Technol. 42 (2), 504e510.
Correia, C.L.T., Gonalves, R.A., Azevedo, M.S., Vieira, M.A., Campos, R.C., 2010.
Determination of total arsenic in seawater by hydride generation atomic uo-
rescence spectrometry. Microchem. J. 96 (1), 157e160.
Corwin, D.L., David, A., Goldberg, S., 1999. Mobility of arsenic in soil from the Rocky
mountain arsenal area. J. Contam. Hydrol. 39 (1), 35e58.
Cummings, D.E., March, A.W., Bostick, B., Spring, S., Caccavo, F., Fendorf, S.,
Rosenzweig, R.F., 2000. Evidence for microbial Fe(III) reduction in anoxic,
mining-impacted lake sediments (Lake Coeur d'Alene, Idaho). Appl. Environ.
Microbiol. 66 (1), 154e162.
Ding, S.M., Han, C., Wang, Y.P., Yao, L., Wang, Y., Xu, D., Sun, Q., Williams, P.N.,
Zhang, C.S., 2015. In situ, high-resolution imaging of labile phosphorus in
sediments of a large eutrophic lake. Water Res. 74, 100e109.
Ding, S.M., Wang, Y., Xu, D., Zhu, C.G., Zhang, C.S., 2013. Gel-based coloration
technique for the submillimeter-scale imaging of labile phosphorus in sedi-
ments and soils with diffusive gradients in thin lms. Environ. Sci. Technol. 47
(14), 7821e7829.
Duan, H.T., Ma, R.H., Xu, X.F., Kong, F.X., Zhang, S.X., Kong, W.J., Hao, J.Y., Shang, L.L.,
2009. Two-decade reconstruction of algal blooms in China's Lake Taihu. Envi-
ron. Sci. Technol. 43 (10), 3522e3528.
Fendorf, S., Michael, H.A., van Geen, A., 2010. Spatial and temporal variations of
groundwater arsenic in South and Southeast Asia. Science 328 (5982),
1123e1127.
Gao, Y., Lesven, L., Gillan, D., Sabbe, K., Billon, G., De Galan, S., Elskens, M.,
Baeyens, W., Leermakers, M., 2009. Geochemical behavior of trace elements in
sub-tidal marine sediments of the Belgian coast. Mar. Chem. 117 (1e4), 88e96.
Gomez-Ariza, J.L., Sa nchez-Rodas, D., Gir aldez, I., Morales, E., 2000. A comparison
between ICP-MS and AFS detection for arsenic speciation in environmental
samples. Talanta 51 (2), 257e268.
Guo, H.M., Stben, D., Berner, Z., 2007. Adsorption of arsenic(III) and arsenic(V) from
groundwater using natural siderite as the adsorbent. J. Colloid Interface Sci. 315
(1), 47e53.
Handley, K.M., McBeth, J.M., Charnock, J.M., Vaughan, D.J., Wincott, P.L., Polya, D.A.,
Lloyd, J.R., 2013. Effect of iron redox transformations on arsenic solid-phase
associations in an arsenic-rich, ferruginous hydrothermal sediment. Geochim.
Cosmochim. Acta 102, 124e142.
Hasegawa, H., 1997. The behavior of trivalent and pentavalent methylarsenicals in
Lake Biwa. Appl. Organomet. Chem. 11, 305e311.
Hasegawa, H., Rahman, M.A., Kitahara, K., Itaya, Y., Maki, T., Ueda, K., 2010. Seasonal
changes of arsenic speciation in lake waters in relation to eutrophication. Sci.
Total Environ. 408 (7), 1684e1690.
He, W., Shang, J.G., Lu, X., Fan, C.X., 2013. Effects of sludge dredging on the pre-
vention and control of algae-caused black bloom in Taihu Lake, China.
J. Environ. Sci. 25 (3), 430e440.
He, Y., Zheng, Y., Locke, D.C., 2007. Cathodic stripping voltammetric analysis of
arsenic species in environmental water samples. Microchem. J. 85 (2), 265e269.
Hossain, M., Williams, P.N., Mestrot, A., Norton, G.J., Deacon, C.M., Meharg, A.A.,
2012. Spatial heterogeneity and kinetic regulation of arsenic dynamics in
mangrove sediments: the Sundarbans, Bangladesh. Environ. Sci. Technol. 46
(16), 8645e8652.
Huerga, A., Lavilla, I., Bendicho, C., 2005. Speciation of the immediately mobilisable
As(III), As(V), MMA and DMA in river sediments by high performance liquid
chromatographyehydride generationeatomic uorescence spectrometry
following ultrasonic extraction. Anal. Chim. Acta 534 (1), 121e128.
Hupfer, M., Dollan, A., 2003. Immobilisation of phosphorus by iron-coated roots of
submerged macrophytes. Hydrobiologia 506e509 (1e3), 635e640.
Jay, J.A., Blute, N.K., Lin, K., Senn, D.B., Hemond, H.F., Durant, J.L., 2005. Controls on
arsenic speciation and solid-phase partitioning in the sediments of a two-basin
lake. Environ. Sci. Technol. 39 (23), 9174e9181.
Liao, M.X., Deng, T., 2006. Arsenic species analysis in porewaters and sediments
using hydride generation atomic uorescence spectrometry. J. Environ. Sci. 18
(5), 995e999.
Ma, R.H., Duan, H.T., Gu, X.H., Zhang, S.X., 2008. Detecting aquatic vegetation
changes in Taihu Lake, China using multi-temporal satellite imagery. Sensors 8
(6), 3988e4005.
 Q. Sun et al. / Environmental Pollution 214 (2016) 314e323
Meharg, A.A., Scrimgeour, C., Hossain, S.A., Fuller, K., Cruickshank, K., Williams, P.N.,
Kinniburgh, D.G., 2006. Codeposition of organic carbon and arsenic in bengal
delta aquifers. Environ. Sci. Technol. 40 (16), 4928e4935.
Moriarty, M.M., Lai, V.W.M., Koch, I., Cui, L.P., Combs, C., Krupp, E.M., Feldmann, J.,
Cullen, W.R., Reimer, K.J., 2014. Speciation and toxicity of arsenic in mining-
affected lake sediments in the Quinsam watershed, British Columbia. Sci. To-
tal Environ. 466, 90e99.
Pedersen, H.D., Postma, D., Jakobsen, R., 2006. Release of arsenic associated with the
reduction and transformation of iron oxides. Geochim. Cosmochim. Acta 70
(16), 4116e4129.
Santner, J., Larsen, M., Kreuzeder, A., Glud, R.N., 2015. Two decades of chemical
imaging of solutes in sediments and soils e a review. Anal. Chim. Acta 878,
9e42.
Qin, B.Q., Xu, P.Z., Wu, Q.L., Luo, L.C., Zhang, Y.L., 2007. Environmental issues of Lake
Taihu, China. Hydrobiologia 581, 3e14.
Qu, W.C., Mike, D., Wang, S.M., 2001. Multivariate analysis of heavy metal and
nutrient concentrations in sediments of Taihu Lake, China. Hydrobiologia 450
(1e3), 83e89.
Rahman, M.A., Hasegawa, H., 2012. Arsenic in freshwater systems: inuence of
eutrophication on occurrence, distribution, speciation, and bioaccumulation.
Appl. Geochem. 27 (1), 304e314.
Root, R.A., Dixit, S., Campbell, K.M., Jew, A.D., Hering, J.G., O'Day, P.A., 2007. Arsenic
sequestration by sorption processes in high-iron sediments. Geochim. Cosmo-
chim. Acta 71 (23), 5782e5803.
Smedley, P.L., Kinniburgh, D.G., 2002. A review of the source, behaviour and dis-
tribution of arsenic in natural waters. Appl. Geochem. 17 (5), 517e568.
Sohrin, Y., Matsui, M., Kawashima, M., Hojo, M., Hasegawa, H., 1997. Arsenic
biogeochemistry affected by eutrophication in Lake Biwa. Jpn. Environ. Sci.
Technol. 31 (10), 2712e2720.
Stockdale, A., Davison, W., Zhang, H., 2009. Micro-scale biogeochemical heteroge-
neity in sediments: a review of available technology and observed evidence.
323
Earth-Sci. Rev. 92 (1e2), 81e97.
Sun, Q., Chen, J., Zhang, H., Ding, S.M., Li, Z., Williams, P.N., Cheng, H., Han, C.,
Wu, L.H., Zhang, C.S., 2014. Improved diffusive gradients in thin lms (DGT)
measurement of total dissolved inorganic arsenic in waters and soils using a
hydrous zirconium oxide binding layer. Anal. Chem. 86 (6), 3060e3067.
Wang, C., Yao, Y., Wang, P., Hou, J., Qian, J., Yuan, Y., Fan, X., 2016. In situ high-
resolution evaluation of labile arsenic and mercury in sediment of a large
shallow lake. Sci. Total Environ. 541, 83e91.
Wang, S., Mulligan, C.N., 2008. Speciation and surface structure of inorganic arsenic
in solid phases: a review. Environ. Int. 34 (6), 867e879.
Wang, S., Cao, X., Lin, C., Chen, X., 2010. Arsenic content and fractionation in the
surface sediments of the Guangzhou section of the Pearl River in Southern
China. J. Hazard. Mater. 183 (1e3), 264e270.
Wei, C.Y., Zhang, N., Yang, L.S., 2011. The uctuation of arsenic levels in lake Taihu.
Biol. Trace Elem. Res. 143 (3), 1310e1318.
Wilkie, J.A., Hering, J.G., 1998. Rapid oxidation of geothermal arsenic(III) in
streamwaters of the eastern Sierra Nevada. Environ. Sci. Technol. 32 (5),
657e662.
Williams, P.N., Santner, J., Larsen, M., Lehto, N.J., Oburger, E., Wenzel, W., Glud, R.N.,
Davison, W., Zhang, H., 2014. Localized ux maxima of arsenic, lead, and iron
around root apices in ooded lowland rice. Environ. Sci. Technol. 48 (15),
8498e8506.
Zhang, C.S., Ding, S.M., Xu, D., Tang, Y., Wong, M.H., 2014. Bioavailability assessment
of phosphorus and metals in soils and sediments: a review of diffusive gradi-
ents in thin lms (DGT). Environ. Monit. Assess. 186 (11), 7367e7378.
Zhang, H., Davison, W., Mortimer, R.J.G., Krom, M.D., Hayes, P.J., Davies, I.M., 2002.
Localised remobilization of metals in a marine sediment. Sci. Total Environ. 296
(1), 175e187.
Zhang, N., Wei, C.Y., Yang, L.S., 2013. Occurrence of arsenic in two large shallow
freshwater lakes in China and a comparison to other lakes around the world.
Microchem. J. 110, 169e177.