Content uploaded by Mirofora Pilakouta
Author content
All content in this area was uploaded by Mirofora Pilakouta on Sep 26, 2018
Content may be subject to copyright.
Vertical distribution of
137
Cs activity concentration in marine
sediments at Amvrakikos Gulf, western of Greece
C. Tsabaris
a
,
*
, D.L. Patiris
a
, E. Fillis-Tsirakis
a
,
b
, V. Kapsimalis
a
, M. Pilakouta
c
,
F.K. Pappa
a
,
b
, R. Vlastou
b
a
Hellenic Centre for Marine Research, Institute of Oceanography, Anavyssos, Greece
b
National Technical University of Athens, Department of Physics, Athens, Greece
c
Technological Education Institute of Piraeus, Automation Engineering Department, Athens, Greece
article info
Article history:
Received 2 December 2014
Received in revised form
28 January 2015
Accepted 7 February 2015
Available online
Keywords:
137
Cs vertical distribution
Marine sediments
1-D model
Mobility
137
Cs
abstract
The aim of the present work is the study of
137
Cs migration in sediment column taking into account the
sedimentation rate in the Amvrakikos Gulf, at the western part of Greece. Marine core sediments were
collected and the measurements were performed using the high resolution gamma-ray spectrometry
method. The vertical distribution of
137
Cs activity concentration, as part of anthropogenic marine
radioactivity, provided averaged sedimentation rate by identifying the depths of activity concentrations
due to the Chernobyl accident and the nuclear tests signals. Furthermore,
137
Cs measurements were
reproduced using the proposed one-dimensional diffusioneadvection model which provides mainly as
an output, the sedimentation rate and the average diffusivity of
137
Cs in the sediment column. The
proposed model estimates the temporal variation of
137
Cs activity concentration from 1987 (one year
after the Chernobyl accident) till today (2014).
©2015 Elsevier Ltd. All rights reserved.
0. Introduction
The radiometric, granulometric and geochemical study of sedi-
ments in coastal areas has great significance in order to understand
the relation between human activities and marine processes. Sed-
iments play an important role in accumulating and transporting
contaminants in the marine environment. Furthermore, the accu-
rate dating of sediments provides significant information about the
dynamic sedimentary processes. These processes require a detailed
sediment geochronology including particulate matter transfer,
sediment mixing and deposition. During the last years a lot of effort
has been made in studying natural and artificial radioactivity in
bottom sediments of closed marine systems (lakes) (Lu and
Matsumoto, 2005; Ligero et al., 2005a; Tsabaris et al., 2007; El-
Reefy et al., 2010) as well as for risk assessment due to the radio-
activity levels of artificial radionuclide,
137
Cs (Külahci and S¸ en,
2009).
The method, which is used widely to determine sedimentation
rates on a timescale of the last century, is the combination of
210
Pb
and
137
Cs activity concentrations in sediment core samples
(Sanchez-Cabeza et al., 1999; Miralles et al., 2005; Alonso-
Hernandez et al., 2006; Lu and Matsumoto, 2005; Tsabaris et al.,
2012). Furthermore, these radionuclides are also utilized as in-
dicators for sediment movement within water bodies, such as open
seas (Sanchez-Cabeza et al., 1999), lakes (Monte et al., 2005b),
rivers (Simms et al., 2008; Monte et al., 2005a) and radiometric
marine applications.
The artificial radionuclide
137
Cs (with half life of 30.05 ±0.08
years) is derived from atmospheric nuclear tests, accidents in nu-
clear plants and authorized discharge of radioactive wastes into the
marine environment.
137
Cs fallout shows: (a) an old peak, which is
attributed to the incidence of nuclear weapons tests, carried out
during the 1950s and 1960s and resulting in a maximum fallout in
1963 (UNSCEAR, 1982; He and Walling, 1996) and (b) a more recent
peak, which corresponds to the Chernobyl accident, in 1986. In
many cases,
137
Cs is hardly observed in sediments due to its low
concentration, the variation of residence time in the water column
and the mobility within sediments caused by diffusion processes
(Ligero et al., 2005a). Furthermore,
137
Cs is one of the main radio-
nuclides which is used as stratigraphic indicator for dating records
*Corresponding author. Hellenic Centre for Marine Research, 46.7 Km Athens-
Sounio Avenue, 19013 Anavyssos, Greece.
E-mail address: tsabaris@hcmr.gr (C. Tsabaris).
Contents lists available at ScienceDirect
Journal of Environmental Radioactivity
journal homepage: www.elsevier.com/locate/jenvrad
http://dx.doi.org/10.1016/j.jenvrad.2015.02.009
0265-931X/©2015 Elsevier Ltd. All rights reserved.
Journal of Environmental Radioactivity 144 (2015) 1e8
since it is incorporated very quickly into the sediment (some
months) depending on the depth of water column. Furthermore,
the diffusivity as well as the settling speed of particulate
137
Cs along
the depth of water columns has been investigated (Tsabaris et al.,
2014).
The selected study area of this work is the Amvrakikos Gulf
which is located at the western part of Greece in Ionian Sea and is
the third biggest gulf of Greece. The gulf is affected by industrial
activities and river discharges. Up to now there is little information
available regarding the
137
Cs radioactivity levels as well as the
sedimentation rates of the specific area. Moreover, no detailed
study has been made for the
137
Cs vertical distribution. Therefore,
the study area is a suitable system for sediment dating and diffu-
sion studies since it is a semi-enclosed system and almost undis-
turbed from tides, waves and currents.
In this work the vertical distribution pattern of
137
Cs was
measured at two locations of the Amvrakikos Gulf. The distribution
of this radionuclide into the sediment column provided dating in-
formation using the two main
137
Cs peaks along with depth.
Moreover, the
137
Cs profiles were theoretically reproduced to pro-
vide information about the mobility of
137
Cs in the sediment col-
umn and more particularly the diffusion and the sedimentation
processes. For this end, a one-dimensional model was applied
based on the vertical distribution of
137
Cs activity concentration in
the sediment column taking into account diffusive and deposition
fluxes. The output of the model was derived analytically and pro-
vided the diffusion coefficient and the sedimentation rate.
1. Materials and methods
1.1. Study area
The Amvrakikos Gulf (Fig. 1) is a semi-enclosed basin of 405 km
2
that receives nutrient inputs, fresh water and sediment fluxes from
a drainage basin of 3850 km
2
(including the catchments of the
rivers Arachthos and Louros) (Friligos et al., 1997). Specifically, the
Louros River contributes with a drainage basin of 865 km
2
and a
mean annual budget of 600 10
6
m
3
y
1
, whereas the Arachthos
River with 1900 km
2
and 2200 10
6
m
3
y
1
, respectively.
In addition, the gulf receives annually through precipitation a
volume of fresh water of approximately 400 10
6
m
3
. The gulf is
relatively shallow (maximum depth of 65 m at the eastern part) and
communicates with the open Ionian Sea through a narrow sill
(Aktion channel of approximately 600 m wide and 5 m depth). The
northern part of the gulf is rather shallow and consists of a rela-
tively wide subaqueous delta/prodelta platform (1e3 km wide),
created by the deposition of sediments from the two main rivers
Arachthos and Louros.
In general, long term sedimentation processes within the gulf
are the combined result of eustatic sea level changes, neotectonic
activity and autocyclidic switching of delta mouths (Kapsimalis
et al., 2005); the latter applies mostly to Arachthos River east-
ward mouth displacement. Recent sedimentation processes affect
the surficial layer (to 40 cm) and the major terrestrial sediment
fluxes are associated with the rivers of the gulf. The Arachthos River
has been reported as the most important sediment source with its
average suspended particulate load to be 7.31$10
6
tonnes a
1
, while
the Louros River discharges a load of <0.5 10
6
tonnes a
1
(Poulos
and Chronis, 1997).
The gulf presents a two-layer water stratification, characterized
by a minimum depth of approximately 5 m, with the upper layer
occupying the first 10 m (except of summer when it is extended up
to 20 m) and the lower layer being rather homogeneous
(Voutsinou-Taliadouri and Balopoulos, 1991). The seasonal varia-
tions of temperature and salinity are similar to those of the open
Ionian Sea and they could be attributed to the difference between
fresh water input (riverine flux and precipitation) and evaporation.
The tidal values are below 20 cm (Tsimplis, 1994), characterized
by a relatively calm wave regime due to the limited wave fetches.
Surface water circulation is also weak, presenting a clockwise trend
(Marinos et al., 1984). Measured current speeds in the central part of
the gulf demonstrate mean values ranging from 0.3 upto 0.19 m s
1
Fig. 1. Study area elocations of core samples.
C. Tsabaris et al. / Journal of Environmental Radioactivity 144 (2015) 1e82
(in winter) (Voutsinou-Taliadouri and Balopoulos, 1991). The same
authors have reported fast currents in the Strait of Preveza, ranging
from 0.12 to 0.15 m s
1
with some bursts reaching 1 m s
1
.
The upper water layer in the gulf is well oxygenated, with dis-
solved oxygen (DO) levels been measured above 7 mg L
1
.In
contrast, the lower layer had values below 3 mg L
1
during spring
and below 1 mg L
1
in summer. Moreover, in a water depth of more
than 40 m and closer to the seabed, the DO was less or equal that
0.3 mg L
1
, indicating the development of anoxic conditions, at least
during the summer period. This pattern is attributed to the strong
stratification of the waters, which inhibits the renewal of deeper
water masses (Poulos et al., 2008). Inter-seasonal measurements,
undertaken in 1987, have revealed the eutrophic character of the
gulf (in particular of Preveza Bay) relative to the surface waters of
the adjacent open Ionian Sea. The increased nutrient concentrations
are also related to high values of chlorophyll-a, with the highest
values measured in spring (22.0e44.8 mg L
1
) and the lowest
during summer and autumn periods (0.3e1.8 mg L
1
)(Gotsis-Skreta
et al., 2000). For comparison, chlorophyll-a levels in the Ionian Sea
are lower than 0.7 mg L
1
, throughout the year (Panayotidis et al.,
1994). More details are given elsewhere (Tsabaris et al., 2011).
In the frame of the IAEA campaign (RER/7/003 project), two
cores were taken for studying the vertical migration of
137
Cs in
locations far from the main river estuaries of the gulf. The 13A core
was selected close to the Aktion channel (where the gulf is con-
nected with the Ionian Sea) and the 13B station was selected in the
middle of the gulf far from the Arachthos and Louros estuaries
(between the eastern and western sub-basins of the gulf). The
maximum depth of the core was 40 cm to focus the study on
137
Cs
depositions during the last century, taking into account typical
values of sedimentation rates.
1.2. Field work and sample analyses
Sediment cores were collected using sediment multicorer Mark
VI (Barnett et al., 1984). Each core was handled by a standard pro-
cedure which includes capping core's ends, storing in vertical po-
sition and refrigerated at 4
C and transporting to the laboratory.
Core samples were prepared at the Institute of Oceanography of the
Hellenic Centre for Marine Research (HCMR), where each core was
split in half and slices of 1e2 cm thick were retrieved. The thickness
resolution in the uppermost part of the core (at a depth of up to
20 cm) was 1 cm and for the lower part of the core (at depths from
20 cm to 40 cm) was 2 cm. Each slice was cleaned for algae and
shells, sieved with a 2 mm sieve and was further divided to grain
size fractions. In the end, the subsamples were placed in special
oven to dry them at 40
C for two days.
137
Cs was measured in sediments at HCMR Environmental
Radioactivity Laboratory using a 50% (nominal relative efficiency)
broad energy HPGe (ORTEC) n-type coaxial detector with a reso-
lution of 1.85 keV at 1.33 MeV (Tsabaris et al., 2011). The energy and
efficiency calibration were performed using a
152
Eu(93%) þ
154
Eu(7%) source that had been prepared and cali-
brated at INP NCSR “Demokritos”according to literature (Tsabaris
et al., 2007, 2012). The prepared reference source was placed in
contact geometry in front of the detector and spectra were recorded
for calculating the efficiency as a function of gamma-ray energy.
Moreover, background (using powder Talc as material in the con-
tainers) and foreground (unknown samples) spectra were acquired
and analyzed for further data deduction.
1.3. Grain size definitions
All samples were split into sand and mud fractions by wet
sieving through a 63
m
m mesh. The coarse-grained subsamples
(diameter >63
m
m) were fractioned by dry sieving, whilst the size
distribution of the fine-grained subsamples were determined by a
grain-size analyzer (Sedigraph Micromeritics 5000 ET). Textural
characterization and grain-size statistics of samples were based on
the classification proposed by Folk (1974), which is widely used by
sedimentologists. According to this classification, the sand fraction
is composed by grains whose diameter varies from 63 to 2000
m
m,
silt with diameters ranging between 4 and 63
m
m, and clay with
diameters less than 4
m
m. A mud fraction is defined the sum of silt
and clay fractions.
2. Model description
The DiffusioneAdvection model is a parabolic differential
equation that describes natural phenomena where particles,
energy or any type of physical quantity of a natural ecosystem is
translated in space, as a result of two basic procedures: diffusion
and advection. In general, the transport of material produces a
deposition flux J
C
¼uC,whereuis the rate of sedimentation and
Cthe activity concentration of the radionuclide under investi-
gation. Furthermore, diffusive transport produces a flux
ðJ
D
¼D$V
x
CÞin the direction of activity concentration gradient
(the direction from layers of greater activity concentration to
those with less activity concentration), where Dis the diffusion
coefficient that takes into account all processes producing
diffusive transport. The diffusion coefficient (D) and the rate of
sedimentation (u) are considered time invariant (Ligero et al.,
2005b). Moreover, in the case of a closed system with no sour-
ces or sink, the one dimension DiffusioneAdvection equation is
described according to Eq. (1):
vC
vt¼DV
2
x
Cðx;tÞuV
x
Cðx;tÞ:(1)
On the right part of Eq. (1), the first term represents the diffusion
and the second the sedimentation process.
The derivation of Eq. (1) is based on the modification of the
continuity equation according to which any variation of a natural
quantity in a natural system is uniquely dependent on its in-and-
out flux. Also, taking any quantity source or sink into account the
continuity equation reads:
vC
vtþV
x
ðD$V
x
Cðx;tÞþuCðx;tÞÞ ¼ s;(2)
where srepresents the quantity source or sink.
In the case of a closed system with no sources or sinks (s¼0),
the above equation leads to the DiffusioneAdvection equation.
Thus, the temporal and spatial variation of the activity con-
centration can be expressed according to Eq. (2), since the vertical
distribution of the activity concentration in marine sediments is,
in fact, for each point an indication of how “long ago”its depo-
sition took place. This model describes how the activity concen-
tration of
137
Cs is changing due to the processes of diffusion and
advection. The radioactive decay will be taken into account later
on, as it decreases the initial concentration of
137
Cs.
The DiffusioneAdvection model describing the vertical dis-
tribution of
137
Cs in soil, as described by Likar et al. (2001) and
applied by Qwasmeh and Fischer (2007), is given according to
the following formula (taking into account the radioactive decay
law):
vC
vt¼Dv
2
C
vx
2
uvC
vxlC:(3)
C. Tsabaris et al. / Journal of Environmental Radioactivity 144 (2015) 1e83
The solution of Eq. (3) according to literature (Likar et al., 2001;
Bossew and Kirchner, 2004) is given according to Eq. (4):
Cðx;tÞ¼C
0Ch
$Gðx;tÞe
lt
þC
0NT
$Gðx;tþt
0
Þe
lðtþt
0
Þ
:(4)
The coefficients C
0Ch
and C
0NT
correspond to the initial apparent
activity concentration of
137
Cs due to the Chernobyl accident and
the nuclear weapon tests, respectively (exactly at the moment of
deposition). The variable xis considered along the vertical axis,
representing depth, while t
0
stands for the time mediating the two
depositions.
Eq. (4) is derived as a linear combination of a Green function G(x,
t) which contains the variations of the
137
Cs activity concentration
due to diffusion and advection processes. The radioactive decay
term “
l
C”in Eq. (3) has turned into the exponential contribution in
both terms of Eq. (4), since decay is the same along each point of the
vertical axis. Radioactive decay can be reasonably considered as a
form of sink for the activity C.
The Green function used in Eq. (4) is given according to Eq. (5):
Gðx;tÞ¼
e
uðxut=2Þ=2D
ffiffiffiffiffiffiffiffiffi
pDt
p
"e
x
2
=
4Dt
u
2ffiffiffiffiffiffi
pt
D
r$e
uðxþut=2Þ=2D
$1Fxþut
ffiffiffiffiffiffiffiffi
4Dt
p#;
(5)
where
F
is the error function for which an approximation was used
according to Eq. (6):
FðxÞ¼ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1exp0
B
@x
2
$
4
p
þax
2
1þax
2
1
C
A
v
u
u
u
u
t;where
a¼8ðp3Þ
3pð4pÞ
z0:140012:
(6)
Significant considerations of this model are the correspon-
dence of the two main initial apparent concentrations of
137
Cs in
the sea bottom (in the years of 1963 and 1986) as
d
-functions. In
fact the advection process moves the centre of each peak for
deeper depths and diffusion spreads the solutes around the
centre of each peak. The solution as given from Eq. (4) in com-
bination with the corresponding expressions of Green function
G(x,t) and error function
F
(x) (as given in Eqs (5) and (6),
respectively) is considered a theoretical prediction derived from
the applied model in order to reproduce the experimental data of
this work.
3. Results
3.1. Grain size
The results of grain size and texture analysis of the uppermost
part of sediment cores are given in Table 1. The core 13A consists of
fine-grained material with relatively increased percentage of sand
(6.02e24.87%). In particular, clay fraction ranges from 33.77% to
54.44%, and silt fraction fluctuates between 35.61% and 46.47%.
According the Folk's nomenclature (Folk, 1974) the sediment
texture of the uppermost sediment (till 40 cm) is characterized as
mud (at depth 2.5e7.5 cm) and sandy mud (at depths 0e1.5 cm and
8.5e40 cm), with the mean size (Mz) varying from 8.8
m
mto
23.4
m
m. The core 13B contains also fine-grained material with silt
and clay percentages exceeding 97%. Clay fraction ranges from 0% to
57.57%, and silt fraction fluctuates between 40.22% and 100%. The
texture in core 13B is characterized as mud for the sediment located
at depth from seafloor to 14.5 cm, and clay at depth from 16.5 cm to
40 cm. The mean size is very low and does not exceed 8.7
m
m.
3.2.
137
Cs profiles
The profile data of the
137
Cs activity concentration for 13A and
13B stations are depicted in Fig. 2a and b, respectively.
137
Cs activity
concentration increases down-core from the surface to the peak
indicating the Chernobyl accident, and then continues to increase
until the lower peak of the nuclear weapon tests. Both sediment
cores were enriched in
137
Cs and their dating is calculated using the
observed peaks of the vertical profiles. The
137
Cs peaks at the 13A
station are identified at 30 and 15.5 cm, respectively while the
maximum activity concentration of
137
Cs was 35 Bq kg
1
. The
137
Cs
peaks in 13B station are identified at 7.5 and 18.5 cm, respectively
while the maximum activity concentration was observed at the
deepest layers of the core. In both cores the
137
Cs contribution from
the weapon tests was more intense compared to the deposition due
to the Chernobyl accident.
The appearance of the
137
Cs peaks (due to nuclear weapon tests
and the Chernobyl accident) gives an estimation of the mean
sedimentation rate in each core assuming constant values for the
diffusion and sedimentation process of
137
Cs along with depth
(Ligero et al., 2005b; Tsabaris et al., 2012). The analysis of the
137
Cs
peaks gives an estimate of the average sedimentation rates and
their uncertainties, which are 0.67 ±0.02 cm y
1
,
0.41 ±0.06 cm y
1
for 13A and 13B cores, respectively.
The overall uncertainties (1
s
) exhibited values 0.02 and
0.06 cm y
1
for the 13A and 13B, respectively. The uncertainty
sources are mainly due to the peak determination and the
compaction effect in the core. The relative uncertainties due to the
peak determination are 2% and 4%, while due to the compaction
effect are 4% and 12% for 13A and 13B stations, respectively. The
relative total uncertainty of the sedimentation rates for 13A and
13B stations is 3 and 14%, respectively.
Table 1
Results of grain size and texture analysis of the sediment cores 13A and 13B.
Sample
name
Depth
(cm)
Sand (%) Silt (%) Clay (%) Mean
size (
m
m)
Texture
(Folk, 1974)
13A 0.5 10.84 35.61 53.55 8.8 sM
1.5 13.07 36.51 50.43 11.7 sM
2.5 8.94 37.14 53.92 12.1 M
3.5 6.02 39.54 54.44 11.1 M
4.5 6.92 39.79 53.29 10.8 M
5.5 6.36 41.48 52.16 11.8 M
6.5 9.58 41.77 48.65 12.4 M
7.5 6.67 46.47 46.86 10.4 M
8.5 16.65 39.74 43.62 16.1 sM
9.5 11.53 40.73 47.73 13.4 sM
20.5 24.87 38.35 36.79 22.8 sM
40.0 20.68 45.54 33.77 23.4 SM
13B 0.5 2.21 40.22 57.57 6.6 M
1.5 2.92 42.18 54.90 8.7 M
2.5 2.16 40.42 57.42 4.8 M
3.5 0.36 42.72 56.92 7.0 M
4.5 1.06 41.99 56.95 7.0 M
5.5 1.21 43.23 55.56 6.0 M
6.5 1.43 43.26 55.31 6.1 M
7.5 1.01 42.06 56.94 4.1 M
8.5 0.73 42.74 56.54 5.3 M
9.5 1.27 44.26 54.47 6.1 M
14.5 0.29 65.51 34.21 6.6 M
16.5 4.01 95.99 0 5.9 Z
18.5 0.37 99.63 0 6.6 Z
22.5 0.41 75.97 23.62 5.8 Z
23.5 0.28 99.72 0 5.8 Z
28.0 0.31 99.70 0 6.9 Z
40.0 0.63 86.93 12.45 6.4 Z
C. Tsabaris et al. / Journal of Environmental Radioactivity 144 (2015) 1e84
3.3. Model results
3.3.1. Model outputs
The model was applied to interpret the profiles of the experi-
mental
137
Cs data implementing the proposed theoretical equation
of advectionediffusion processes. The solution of Eq. (6) provides
the theoretical spatial (one dimensional) and temporal variation of
137
Cs activity along with depth taking into account the diffusion
and mobility processes as well as the apparent initial deposition of
137
Cs on the seabed from the Chernobyl accident and nuclear
weapon tests. The model assumes two
d
-Dirac contributions
exactly at 1963 and 1986, which they change in Gaussian-shaped
form along with time (which subsequently appears as a broad
peak at deeper depths). The results of the model are depicted in
Fig. 2a and b for 13A and 13B, respectively.
The agreement between experimental and theoretical data is
satisfactory along with depth at 13A position. At the surficial
sediment layers the activity concentration is low (from 0 to 5 cm
depth there is a slight variation of
137
Cs activity concentration). The
contribution from fallout due to the Chernobyl accident is repre-
sented by two folded contributions at 10 and 15 cm (possibly due to
the different hydrological processes of
137
Cs accumulation into the
sediment after 1986). Such processes could be attributed to annual
hydrological and meteorological conditions in the study area. The
model provides as output the initial apparent activity concentra-
tion of
137
Cs at the studied area (at 1963 and 1986, respectively), an
estimation of sedimentation rate and the vertical molecular diffu-
sion of
137
Cs into the sediment. The fitting process of the theoretical
solution to the experimental data at 13A position provided the
following activity concentrations in the sediment after the Cher-
nobyl accident and nuclear tests: (98 ±10) B q kg
1
and
(420 ±40) Bq kg
1
, respectively. Moreover, the diffusivity and
vertical velocity (sedimentation rate) are (0.21 ±0.02) cm
2
y
1
and
(0.64 ±0.05) cm y
1
, respectively.
As concerns the location 13B, the experimental data are in a very
good agreement with the theoretical data, only from surface to
25 cm depth. At deeper depths (>25 cm), the estimated data do not
reproduce the experimental profile. This difference could be
attributed to the dependence of the diffusion and deposition fluxes
in the periods that correspond to depths below 25 cm. The
contribution from Chernobyl accident and nuclear tests are also
reproduced at 8 cm and at 18 cm, respectively. The
137
Cs activity
concentration after the Chernobyl accident and nuclear tests are
(60 ±5) Bq kg
1
and (211 ±18) Bq kg
1
, respectively. Moreover, the
effective diffusivity and sedimentation rate are
(0.10 ±0.02) cm
2
y
1
and (0.41 ±0.04) cm y
1
, respectively.
3.3.2. Temporal vertical profiles
The outputs of the model were used for providing past and
future trends of the
137
Cs distribution as a function of depth. The
evaluation of the distribution is estimated at two different dates:
the year of the sampling (2007) and the present year (2014). The
estimated data are depicted in Fig. 3 and represent the temporal
evolution along with the depth of the core 13A taking into account
that there was no new
137
Cs input in the closed system of the gulf
(except the two fallouts). Assuming that the studied areas are
closed systems, the estimated values in 2014 exhibit broader con-
tributions (along with depth) compared with the distribution in
2007. Furthermore, the maximum values of the peaks (12 Bq kg
1
of the Chernobyl contribution and 22 Bq kg
1
of the nuclear
weapon tests contribution) at 2014 are reduced compared with the
peak values at 2007 (17 Bq kg
1
of the Chernobyl contribution and
28 Bq kg
1
of the nuclear weapon tests contribution).
4. Discussion
The Amvrakikos Gulf is a semi-enclosed system almost undis-
turbed from tides, strong waves and currents. The applied model
reproduced the vertical distribution of
137
Cs, the average sedi-
mentation rate and the average diffusion coefficient values
assuming that both of them are independent of time. Furthermore,
previous studies based on sediment cores description and high
resolution seismic reflection profiles have not identified any
recently triggered submarine mass-movements in the gulf
(Anastasakis et al., 2007; Ferentinos et al., 2010; Kapsimalis et al.,
2005; Naeher et al., 2012; Poulos et al., 1995, 2008). In addition,
sediment resuspension is unlikely to occur in the basin, since bot-
tom currents are very weak with speeds less than 1 m s
1
(Balopoulos and Papageorgiou, 1989; Kountoura and Zacharias,
Fig. 2. (a) The
137
Cs activity concentration along with depth (location 13A). The
experimental data are depicted as circles and model estimations as solid line. (b) The
137
Cs activity concentration along with depth (location 13B). The experimental data are
depicted as circles and model estimations as solid line.
C. Tsabaris et al. / Journal of Environmental Radioactivity 144 (2015) 1e85
2011). Thus, in this work occasional deposition processes (i.e.,
reworking and resuspension) are considered as negligible.
The proposed model attributes the mobility of
137
Cs in the
sediment column taking into account its vertical transport gener-
ated by the sedimentation and the diffusion processes. The average
sedimentation rates are compared with other estimates using the
vertical distribution of
210
Pb in sites close to 13A and 13B sampling
stations. More specifically, in the vicinity of 13A site (very close to
Aktion channel) comparable average values were estimated by
Naeher et al. (2012) (0.6 cm y
1
) and Eleftheriou (2014)
(0.64 ±0.10 cm y
1
). The
210
Pb profile analysis of 13B core by
Eleftheriou (2014) showed an average sedimentation rate of
0.36 ±0.03 cm y
1
which is in a good agreement (within
uncertainties) with the calculated value in the present study
(0.41 ±0.06 cm y
1
). The lower values of the sedimentation rate in
13B station compared to those of 13A station may be attributed to
the different geomorphological characteristic of these two sites,
since 13B is located between the eastern and western sub-basins of
the gulf, and thus is not strongly affected by the main rivers.
As concerns the diffusion process which is responsible for the
migration of
137
Cs in the sediment column, the molecular diffusion
of the radionuclide in the interstitial water of the sediment is
studied taking into account the distribution of the radionuclide
between sedimentary medium and liquid phase. If the radionuclide
is subject to linear sorption process, then Dis scaled by a retarda-
tion factor, R
d
(Flury and Gimmi, 2002):
R
d
¼1þr
3
K
d
(7)
where,
r
is the density of the sediment, 3the porosity and K
d
is the
distribution coefficient of concentrations in the interface between
solid and liquid phases (mass activity concentration in sediment/
activity concentration in seawater). Taking into account the
freedom of migration of the radionuclide in the interstitial water,
the coefficient of molecular diffusion is given by the expression
(Flury and Gimmi, 2002):
D
molec
¼3
2
D
Cs
R
d
;(8)
where, D
Cs
is the coefficient of molecular diffusion of
137
Cs in pore
water (Li and Gregory, 1974) which is approximately 10
5
cm
2
s
1
.
As concerns the 13A core samples, the average porosity is 40%,
the average wet density of the samples is 1 g cm
3
and the distri-
bution coefficient, K
d
is (1000 ±170) L k g
1
(which is given by the
ratio of the activity
137
Cs concentration of the surficial sediment
2.9 ±0.4 Bq kg
1
divided by the activity concentration of seawater
at the sea-bottom interface 2.9 ±0.3 Bq m
3
). The molecular
diffusion, D
molec
, for 13A core is calculated from Eq. (8) and is equal
(6.4 ±1.0) 10
9
cm
2
s
1
or (0.20 ±0.03) cm
2
y
1
. In respect of the
diffusion coefficient for
137
Cs (D) in 13A core samples, its value
(0.20 ±0.03 cm
2
y
1
) is in satisfactory agreement with the one
determined by the proposed model of this work
(0.21 ±0.02) cm
2
y
1
). The reproduction of the estimated diffusivity
by Eq. (8) demonstrates that the main process of the vertical
migration of
137
Cs is diffusion though the pore water in the sedi-
mentary medium.
As concerns the 13B core samples, the average porosity is 50%,
the average wet density of the samples is 1 g cm
3
and the distri-
bution coefficient, K
d
, is (3600 ±540) L kg
1
(given by the ratio of
the activity
137
Cs concentration of the surficial sediment
(11 ±1) Bq kg
1
divided by the activity concentration in the bottom
seawater 3.1 ±0.4 Bq m
3
). The molecular diffusion, D
molec
, for 13A
core is calculated from Eq. (8) and is (3.8 ±0.5) 10
9
cm
2
s
1
or
(0.12 ±0.02) cm
2
y
1
. The molecular coefficient diffusion for
137
Cs in
13B core is overestimated by 20% compared with the estimated
results of the applied model. This difference may be attributed to
the fact that part of the diffusion comes from the mixture of benthic
biological organisms with the sediment material (Ligero et al.,
2005b) or due to the organic materials observed in the vicinity of
the studied stations (Naeher et al., 2012).
Finally, the diffusion coefficient value of 13B station was found
50% lower compared to the data of 13A station. This could be
attributed mainly to the different sediment texture of the samples.
Fine grain sediment is more abundant in case of 13B station
resulting in more effective
137
Cs molecules sorption and the diffu-
sion flux is significantly reduced.
5. Conclusions
In this work the sedimentation process is studied taking into
account the vertical mobility of
137
Cs in the core sediment following
the diffusion and sedimentation processes. The fitting procedure of
the experimental data using the proposed theoretical solution
provided sedimentation rates that are in a good agreement with the
experimental results. In fact the rate of sedimentation as deduced
from the model reproduces the experimental values. More
specifically:
The model reproduced the experimental data satisfactorily for
13A core. The model provides estimated results of
137
Cs activity
concentration for two peaks (one due to the nuclear weapon
tests from 1963 and another due to Chernobyl accident in 1986).
The output of the model is the apparent initial activity
Fig. 3. The estimated
137
Cs activity concentration along with depth for 13A station is
represented in the year of sampling as well in the present year (2007, 2014).
C. Tsabaris et al. / Journal of Environmental Radioactivity 144 (2015) 1e86
concentration of
137
Cs for both contributions, the average sedi-
mentation rate and the diffusion coefficient.
The model reproduced partially the vertical distribution of the
137
Cs activity concentration for 13B station. The reproduced data
are in good agreement above 25 cm, while discrepancies of
10e20% were observed at depths below 25 cm. This difference
may attributed to inefficiency of the model to reproduce folded
contributions (observed as tail at the deep layer of the core).
These contributions are due to the steadily inputs during the
long period of nuclear weapon tests (from 1954 till 1963).
The 13A station is close to the Aktion channel and the average
diffusivity is almost 2 times higher compared with the 13B station.
This difference is attributed to the different sediment texture along
with depth between the studied sediment cores since the sorption
of the
137
Cs molecules is more efficient onto finer grain sediment
which decreases the diffusion flux (Ligero et al., 2005b).
The
137
Cs activity concentration is estimated along with depth in
different years (2007, 2014) according to the main signals of
137
Cs
deposition (Chernobyl accident and nuclear weapon tests). The
average sedimentation rate was reproduced as well as the diffusion
coefficient from an empirical model (Ligero et al., 2005b) applied in
the sedimentewater interface. In future other parameters (e.g.
organic matter, mineralogy) will be taken into account for studying
the spatial and temporal behaviour of Cs in the sedimentary
medium.
Acknowledgements
The authors would like to thank the International Atomic Energy
Agency for the financial support of the campaign in the frame of the
RER/7/003, as well as Dr I. Osvath for supporting this work. Addi-
tionally, the authors would like to thank the staff of the research
vessel “Palagruza”and the organizers for the technical support
during the campaign.
References
Alonso-Hernandez, C.M., Diaz-Asencio, M., Munoz-Caravaca, A., Delfanti, R.,
Papucci, C., Ferretti, O., Crovato, C., 2006. Recent changes in sedimentation
regime in Cienfuegos Bay, Cuba, as inferred from
210
Pb and
137
Cs vertical pro-
files. Cont. Shelf Res. 26, 153e167.
Anastasakis, G., Piper, D.J.W., Tziavos, C., 2007. Sedimentological response to neo-
tectonics and sea-level change in a delta-fed, complex graben: Gulf of Amvra-
kikos, western Greece. Mar. Geol. 236, 27e44.
Balopoulos, E., Papageorgiou, E., 1989. Physical oceanographic characteristics and
sea currents in Amvrakikos Gulf (Eastern Ionian Sea). In: Tsiavos, Ch (Ed.),
Oceanographic Study of the Amvrakikos Gulf. Physical Oceanography. Final
Report, vol. 1. Athens, Greece (in Greek).
Barnett, P.R.O., Watson, J., Connelly, D., 1984. A multiple corer for taking virtually
undisturbed samples from shelf, bathyal and abyssal sediments. Ocean. Acta 7,
399e408.
Bossew, P., Kirchner, G., 2004. Modelling the vertical distribution of radionuclides in
soil. Part 1: the convectionedispersion equation revisited. J. Environ. Radioact.
73, 127e150.
Eleftheriou, G., 2014. Estimation and SpatialeTemporal Dispersion of Radionuclides
in the Aquatic Environment (PhD thesis). National Technical University of
Athens.
El-Reefy, I.H., Sharshar, T., Elnimr, T., Badran, M.H., 2010. Distribution of gamma-ray
emitting radionuclides in the marine environment of the Burullus lake: II.
Bottom sediments. Environ. Monit. Assess. 169, 273e284.
Ferentinos, G., Papatheodorou, G., Geraga, M., Iatrou, M., Fakiris, E.,
Christodoulou, D., Dimitriou, E., Koutsikopoulos, C., 2010. Fjord water circula-
tion patterns and dysoxic/anoxic conditions in a Mediterranean semi-enclosed
embayment in the Amvrakikos Gulf, Greece. Estuar. Coast. Shelf Sci. 88,
473e481.
Flury, M., Gimmi, T., 2002. Solute diffusion. In: Dane, J.H., Topp, G.C. (Eds.), Methods
of Soils Analysis, Part 4, Physical Methods. Soil Science Society of America,
Madison, WI, pp. 1323e1351.
Folk, R.L., 1974. Petrology of the Sedimentary Rocks. Hemphill Publishing Company,
Austin, Texas, p. 182.
Friligos, N., Balopoulos, E.T., Psyllidou-Giouranovits, R., 1997. Eutrophication and
hydrography in the Amvrakikos Gulf, Ionian Sea. Fresenius Environ. Bull. 6,
21e26.
Gotsis-Skreta, O., Psochiou, E., Lempesis, G., Brampa, D., Theodorou, D.,
Balopoulos, E., 2000. Seasonal variation of phytoplankton and environmental
parameters in a semi-enclosed sea basin, Amvrakikos Gulf, Ionian Sea. In: Proc.
6th Hellenic Symposium on Oceanography and Fisheries, vol. 1, pp. 512e516.
Chios, Greece.
He, Q., Walling, D.E., 1996. Interpreting particle size effects in adsorption of
137
Cs
and unsupported
210
Pb by mineral soils and sediments. J. Environ. Radioact. 30,
117e137.
Kapsimalis, V., Pavlakis, P., Poulos, S.E., Alexandri, S., Tziavos, C., Sioulas, A.,
Filippas, D., Lykousis, V., 2005. Internal structure and evolution of the Late
Quaternary sequence in a shallow embayment: the Amvrakikos Gulf, NW
Greece. Mar. Geol. 222e223, 399e418.
Kountoura, K., Zacharias, I., 2011. Temporal and spatial distribution of hypoxic/
seasonal anoxic zone in Amvrakikos Gulf, western Greece. Estuar. Coast. Shelf
Sci. 94, 123e128.
Külahci, F., S¸ en, Z., 20 09. Risk assessment of distribution coefficient from
137
Cs
measurements. Environ. Monit. Assess. 149, 363e370.
Li, Y.H., Gregory, S., 1974. Diffusion of ions in sea water and deep-sea sediments.
Geochim. Cosmochim. Acta 38, 703e714.
Ligero, R.A., Barrera, M., Casas-Ruiz, M., 2005a. Levels of
137
Cs in muddy sediments
of the seabed of the bay of Cadiz, Spain. Part I. Vertical and spatial distribution
of activities. J. Environ. Radioact. 80, 75e86.
Ligero, R.A., Barrera, M., Casas-Ruiz, M., 2005b. Levels of
137
Cs in muddy sediments
of the seabed of the bay of Cadiz, (Spain). Part II. Model of vertical migration of
137
Cs. J. Environ. Radioact. 80, 87e103.
Likar, A., Omahen, G., Lipoglav
sek, M., Vidmar, T., 2001. A theoretical description of
diffusion and migration of
137
Cs in soil. J. Environ. Radioact. 57, 191e201.
Lu, X., Matsumoto, E., 2005. Recent sedimentation rates derived from
210
Pb and
137
Cs methods in Ise Bay, Japan. Estuar. Coast. Shelf Sci. 65, 83e93.
Marinos, G., Marinos, P., Stournaras, G., Karotsieris, Z., 1984. A Detailed Study on the
Combined Utilisation of the Water Resources of the Rivers Louros and Arach-
thos Drainage Basins. Technical report to the Ministry of Energy and Natural
Resources, Athens.
Miralles, J., Radakovitch, O., Aloisi, J.-C., 2005.
210
Pb sedimentation rates from the
Northwestern Mediterranean margin. Mar. Geol. 216, 155e167.
Monte, L., Boyer, P., Brittain, J.E., Håkanson, L., Lepicard, S., Smith, J.T., 2005a. Review
and assessment of models for predicting the migration of radionuclides through
rivers. J. Environ. Radioact. 79, 273e296.
Monte, L., Grimani, C., Desideri, D., Angeli, G., 20 05b. Modelling the long-term
behaviour of radiocesium and radiostrontium in two Italian lakes. J. Environ.
Radioact. 80, 105e123.
Naeher, S., Geraga, M., Papatheodorou, G., Ferentinos, G., Kaberi, H., Schubert, C.J.,
2012. Environmental variations in a semi-enclosed embayment (Amvrakikos
Gulf, Greece) ereconstructions based on benthic foraminifera abundance and
lipid biomarker pattern. Biogeosciences 9, 5081e5094.
Panayotidis, P., Pancucci, M.A., Balopoulos, E., Gotsis-Skretas, O., 1994. Plankton
distribution patterns in a Mediterranean dilution basin: Amvrakikos Gulf
(Ionian Sea, Greece). Mar. Ecol. 15, 93e104.
Poulos, S.E., Lykousis, V., Collins, M.B., 1995. Late quaternary deltaic evolution of
Amvrakikos Gulf, western Greece. Geo Mar. Lett. 15, 9e16 .
Poulos, S., Chronis, G.T., 1997. The importance of the Greek river systems in the
evolution of the Greek coastline. In: Briand, F., Maldonado, A. (Eds.), Trans-
formations and Evolution of the Mediterranean Coastline, CIESM Science Series
No. 3. Bulletin de l'Institut Oceanographique, Monaco, pp. 75e96. No. 18.
Poulos, S.E., Kapsimalis, V., Tziavos, C., Paramana, T., 2008. Origin and distribution
of surface sediments and human impacts on recent sedimentary processes.
The case of the Amvrakikos Gulf (NE Ionian Sea). Cont. Shelf Res. 28,
2736e2745.
Qwasmeh, A., Fischer, H.W., 2007. Analysis of the vertical migration of
137
Cs in
Jordanian soils using the diffusioneconvection model. In: Book of Extended
Synopses of International Conference of Environmental Radioactivity: from
Measurements and Assessments to Regulation, pp. 295e296. Vienna, Austria.
Sanchez-Cabeza, J.A., Masque, P., Ani-Ragolta, I., Merino, J., Frignani, M., Alvisi, F.,
Palanques, A., Puig, P., 1999. Sediment accumulation rates in the southern
Barcelona continental margin (NW Mediterranean Sea) derived from
210
Pb and
137
Cs chronology. Prog. Oceanogr. 44, 313e332.
Simms, A.D., Woodroffe, C., Jones, B.G., Heijnis, H., Mann, R.A., Harrison, J., 2008. Use
of
210
Pb and
137
Cs to simultaneously constrain ages and sources of post-dam
sediments in the Cordeaux reservoir, Sydney, Australia. J. Environ. Radioact.
99 , 1111 e1120 .
Tsabaris, C., Eleftheriou, G., Kapsimalis, V., Anagnostou, C., Vlastou, R., Durmishi, C.,
Kedhi, M., Kalfas, C.A., 2007. Radioactivity levels of recent sediments in the
Butrint Lagoon and the adjacent coast of Albania. J. Appl. Radiat. Isot. 65,
445e453.
Tsabaris, C., Evangeliou, N., Fillis-Tsirakis, E., Sotiropoulou, M., Patiris, D.L., Florou, H.,
2011. Distribution of natural radioactivity in sediment cores from Amvrakikos
Gulf (western Greece) as part of IAEA's campaign in Adriatic and Ionian Seas.
Radiat. Prot. Dosim. 150, 474e487.
Tsabaris, C., Kapsimalis, V., Eleftheriou, G., Laubenstein, M., Kaberi, H., Plastino, W.,
2012. Determination of
137
Cs activities in surface sediments and derived sedi-
ment accumulation rates in Thessaloniki Gulf, Greece. Environ. Earth Sci. 67,
833e843.
C. Tsabaris et al. / Journal of Environmental Radioactivity 144 (2015) 1e87
Tsabaris, C., Zervakis, V., Kaberi, H., Delfanti, R., Georgopoulos, D.,
Lampropoulou, M., Kalfas, C.A., 2014.
137
Cs vertical distribution at the deep
basins of the North and Central Aegean Sea, Greece. J. Environ. Radioact. 132,
47e56.
Tsimplis, M.N., 1994. Tidal oscillations in the Aegean and Ionian seas. Estuar. Coast.
Shelf Sci. 3, 201e208.
UNSCEAR, 1982. Ionising Radiation: Sources and Biological Effects. A report to the
General Assembly. United Nations, New York.
Voutsinou-Taliadouri, F., Balopoulos, E.T., 1991. Geochemical and physical oceano-
graphic aspects of the Amvrakikos Gulf (Ionian Sea, Greece). Toxicol. Environ.
Chem. 31e32, 177e18 5.
C. Tsabaris et al. / Journal of Environmental Radioactivity 144 (2015) 1e88
