Content uploaded by Helen Papaefthymiou
Author content
All content in this area was uploaded by Helen Papaefthymiou on Dec 01, 2015
Content may be subject to copyright.
Natural radionuclides and
137
Cs distributions and
their relationship with sedimentological processes
in Patras Harbour, Greece
H. Papaefthymiou
a,
*, G. Papatheodorou
b
, A. Moustakli
a
,
D. Christodoulou
b
, M. Geraga
b
a
Division of Physical, Inorganic and Nuclear Chemistry, Department of Chemistry,
University of Patras, Patras 265 04, Greece
b
Laboratory of Marine Geology and Physical Oceanography, Department of Geology,
University of Patras, Patras 265 04, Greece
Received 8 June 2006; received in revised form 20 December 2006; accepted 22 December 2006
Available online 26 February 2007
Abstract
Surficial and subsurficial sediment samples derived from gravity cores, selected from the harbour of
Patras, Greece, were analyzed for grain size, water content, bulk density, specific gravity, organic carbon
content and specific activities of natural radionuclides and
137
Cs. The specific activities of
232
Th,
226
Ra,
40
K and
137
Cs were measured radiometrically. The radionuclides
238
U and
232
Th were also analyzed using
the INAA. The differences found between the specific activities of the natural radionuclides measured by
the two methods are of no statistical significance. The sediment cores selection was based on a detailed
bathymetric and marine seismic survey. Through the study of the detailed bathymetric map and the seis-
mic profiles it was shown that ship traffic is highly influential to the harbour bathymetry. The granulomet-
ric and geotechnical properties of the sediments and therefore the specific activities of the natural
radionuclides and
137
Cs seem to be controlled by the ship traffic. Relationship between radionuclide
activity concentrations and granulometric/geotechnical parameters was defined after the treatment of all
the analyses using R-mode factor analysis. The natural radionuclide activities are related to the fine frac-
tion and bulk density of the sediments, while
137
Cs is mainly influenced by the organic carbon content. In
* Corresponding author. Tel.: þ30 2610997132; fax: þ30 2610997118.
E-mail addresses: epap@chemistry.upatras.gr (H. Papaefthymiou), george.papatheodorou@upatras.gr (G. Papatheodorou).
0265-931X/$ - see front matter Ó2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jenvrad.2006.12.014
Journal of Environmental Radioactivity 94 (2007) 55e74
www.elsevier.com/locate/jenvrad
addition,
238
U and
226
Ra seem to be in close relation with the heavy minerals fraction in coarse-grained
sediments with high specific gravity.
Ó2007 Elsevier Ltd. All rights reserved.
Keywords: Natural radioactivity;
137
Cs; Marine sediments; Harbour; Multivariate analysis; Greece
1. Introduction
For several decades, the occurrence of natural and anthropogenic radionuclides in the marine
environment has been a subject of study, in order to better understand the natural processes both
in the seafloor and the water column as well as the protection of the marine environment.
The natural radionuclides are incorporated as trace elements in the crystal lattice of the min-
erals, and are present in the sediments with their activities being dependent on the type of the
minerals (Dunk et al., 2002). The radionuclides, as natural constituents of the marine environ-
ment, are derived from the radionuclide-bearing rocks erosion. As a result of the weathering
and erosive processes, the minerals containing radionuclides are transported to the marine
environment and eventually are incorporated into the seafloor sediments. In addition to the nat-
ural inputs of the radionuclides, there are also major anthropogenic sources, including waste
discharges from phosphate fertilizer plants and heavy waters produced in reactor cooling sys-
tem in nuclear power plants. Anthropogenic radionuclides, such us
137
Cs are also derived from
the following sources: (i) the fission and activation products created from nuclear testing and
deposited atmospherically, with a maximum discharge in 1958 and 1963, and (ii) the Chernobyl
accident which produced a peak discharge in 1986 (Cundy and Croudace, 1995).
In the marine environment,
137
Cs and natural radionuclides, mainly
238
U and its daughters, are
deposited on the seafloor through a wide range of processes, including fixation on suspended
particulate matter and sedimentation, direct precipitation of colloidal forms, absorption on
clay minerals and complexion with organic material (Ligero et al., 2001; Park et al., 2004). In
marine settings such as harbours, there may be a dramatic change in these processes as a result
of human impact on the system. Siltation is of great interest in many harbours. In such cases
dredging has a major impact upon the distribution of sediments. The introduction of high-speed
ferries with hydro-jet propulsion led to the erosion of surface sediments in many harbours.
As a primary objective of this paper we set the investigation of the spatial and temporal
distribution patterns of natural radionuclides and
137
Cs in the sediments of the Patras Harbour.
We also attempt to elucidate the sedimentological processes affecting the distribution of radio-
nuclides in an environment influenced to a great extent by human activities. Using multivariate
statistical methods we studied the relationship between the aforementioned activities and some
granulometric/geotechnical parameters. So far our knowledge of the radionuclide concentration
levels in the marine environment of Greece is very limited (Florou and Kritidis, 1991; Papa-
theodorou et al., 2005) and thus a second but no less important aim of this work is to provide
a useful baseline data set for the natural radionuclides and
137
Cs levels for future research.
2. Study area
Patras Harbour is located along the south-eastern coast of the Gulf of Patras, in Western
Greece (Fig. 1). The centre of the development and expansion of the city of Patras is the
56 H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
Fig. 1. Detailed bathymetric map of Patras Harbour showing the three harbour basins (A, B and C), the sediment coring
stations, the vessel main routes and the turning sites (tp). Continuous lines indicate 3.5 kHz profiles showed in Fig. 2.
Isobaths were constructed by kriging method using GIS (T1eT5: terminals, D1eD4: docks).
57H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
harbour. As a result of its proximity to Patras city (200 000 citizens), the harbour is contami-
nated due to shipping and urban activities. The harbour gained its economic significance in
the early 20th century as it was the only communication and commercial link between Western
Greece and Italy. Today, Patras Harbour is classified among the leading ports in Greece on the
basis of tourists’ traffic.
The harbour covers an area of about 0.6 km
2
with an average depth of 10.5 m and consists of
three bowl shaped basins (A, B and C), which are protected by a NeS trending breakwater
(Fig. 1). Erosion of the surface sediments is prevalent in areas with heavy ship traffic especially
the high-speed ferries from Patras to Italy. On the other hand, undisturbed seafloor is found in
areas with low ship traffic (Fig. 1).
The harbour and the city of Patras were founded over the Holocene deposits and the Plio-
Pleistocene sediments. The Plio-Pleistocene sediments constitute the geological base both of
the harbour and the city of Patras (Koukis et al., 2005). Based on marine seismic data, the
Plio-Pleistocene sediments are covered by a thick sequence (20 m) of Holocene fine-grained
sediments in the littoral zone of Patras city (Papatheodorou et al., 1993; Hasiotis et al.,
1996; Christodoulou et al., 2003).
3. Materials and methods
3.1. Field work
The data used in this survey are based largely upon the marine remote sensing and sedimentological/
geotechnical surveys carried out to support dredging projects in Patras Harbour. These surveys were
conducted using an ODOM ECHOTRAC DF 3200 MKII 200 kHz echo sounder and a 3.5 kHz subbottom
profiling system. The sampling stations were planned on the basis of the results of these surveys. Six sed-
iment cores in total (74 mm in diameter) were collected by a 3-m-long Benthos gravity corer (Fig. 1) and
the core positioning was fixed with a differential GPS with a RMS accuracy of about 1 m. All sediment
cores were collected from areas of sedimentation and more or less undisturbed seafloor. None of the core
was selected from areas of erosion. Therefore, the spatial sampling coverage was not homogeneous in the
harbour.
3.2. Laboratory tests and analyses
The retrieved 50e150 cm long sediment cores were stored vertically at 4 C. The gravity cores were
cut in half gently with a Plexiglas sheet and were described according to colour and textural units. The
uppermost 2 cm of all cores were used for further analysis along with selected downcore sub-samples,
2 cm thick. The sub-samples were selected on the basis of the cores macroscopic description.
Granulometric analysis was carried out using standard sieve and pipette techniques after the destruc-
tion with H
2
O
2
of the organic matter in all sub-samples. The evaluation of the sediment texture and sta-
tistical grain size parameters was effectuated according to Folk (1974). The organic carbon content (C
org
)
(%) was determined by oxidation with 1 N K
2
Cr
2
O
7
, acidified with concentrated H
2
SO
4
and titration with
0.5 N Fe (NH
4
)
2
(SO
4
)
2
. All sub-samples were subjected to water content (%), bulk density (g/cm
3
) and
specific gravity measurements using standard methods (BSI, 1975).
Two sub-samples from each sediment core, the surface and the deepest one, were used for the deter-
mination of the specific activities of
226
Ra,
232
Th,
40
K and
137
Cs by high-resolution g-ray spectrometry,
while three or four sub-samples from each sediment core were used for the determination of uranium
and thorium concentrations by INAA. Prior to analyses, the collected samples were air dried, crushed
to fine powder and homogenized. In order to perform the specific activity measurements, the homogenized
samples were dried at 105 C to constant weight, packed into cylindrical containers, weighted, sealed
58 H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
hermetically and stored for a minimum period of 4 weeks prior to counting to establish secular radioactive
equilibrium between
226
Ra and its short-lived daughter products. Both energy and efficiency calibration
procedures and activity concentration measurements are described in detail elsewhere (Papaefthymiou
et al., in press). Briefly, all countings were performed using a Canberra HPGe detector with an energy
resolution of 1.9 keV at the 1.33 MeV of
60
Co g-ray and relative efficiency 25%. The minimum counting
time was 24 h. The spectra were analyzed by using the GAMANAL routine, included in the GANAAS
package, distributed by IAEA, while peak areas were determined using Ganaas 3.1 computer program
of IAEA. The accuracy was estimated against the IAEA-312 and IAEA-375 reference materials. The tests
showed good correspondence with the certified/recommended values (<5%).
Concentrations of uranium and thorium were also measured by Instrumental Neutron Activation Anal-
ysis (INAA) in combination with the high-resolution g-ray spectrometry. The detailed procedure is
described elsewhere (Papaefthymiou et al., 2005). The accuracy of the method was evaluated on the basis
of the analysis of the standards: Soil-7 and SL-1 from IAEA and 1633b Fly-ash from NIST by the same
procedure. The relative uncertainties of the certificated/recommended values ranged from 6 to 8%. The
detection limits for the determination of uranium and thorium by INAA were 0.5 mg kg
1
. All countings
were performed at the Nuclear Chemistry Laboratory, Department of Chemistry of the University of
Patras. In order to compare them with the direct spectrometry measurements, the concentrations (mg kg
1
)
of uranium and thorium were converted into activity concentrations (Bq kg
1
).
3.3. Statistical treatment
The relationship between radionuclide activities and several sedimentological/geotechnical parameters
was examined using multivariate analysis. More specifically, it was used the R-mode factor analysis, as the
results it provides are more useful than the descriptive statistics (Reyment and Joreskog, 1996). It is also
widely applied among others in marine geochemistry (Papatheodorou et al., 1999, 2002).
In this study, the multivariate statistical method was applied in two separate data sets. The first data set
(9 20) was constructed by 20 sediment samples in total, which have been analyzed for
238
U and
232
Th,
as well as for organic carbon content, sand, <63 mm fraction, mean grain size (f), water content (%), bulk
density and specific gravity. Twelve of the above-mentioned samples, which have been further analyzed
for
226
Ra,
40
K and
137
Cs constituted the second data set (12 12) resulting in a 12 12 matrix.
Normality tests showed that the data distributions of the examined variables are not normal. The vari-
able values of raw data establish the absolute values of kurtosis ranging from 1.25 to 9.02 and skewness
values varying from 0.58 to 2.89, suggesting that the variables do not have normal populations. Thus, the
raw data of all variables were transformed (log x) and standardized according to Reimann and Filzmoser’s
(2000) method. All variables’ values for the transformed data are closer to normal ranging from 0.93 to
0.96 except for the bulk density value.
In order to study in detail the connection between radionuclide activities and granulometric/geotech-
nical parameters from both data sets it was employed a ‘‘combined’’ R-mode factor analysis. A first order
factor analysis was carried out on the first data set (9 20), while a three-factor model was chosen to
explain the interrelationships among the variables of the first data set. The selection of the number of
the statistically significant factors was based on a combination of criteria (Papatheodorou et al., 2006).
The interpretation of these factors was based on known information regarding the chemical phases of
the marine sediments. The calculated factor scores of each factor represent the ‘‘amounts’’ of each process
in every sample, given that the factors can be considered as processes combining radionuclides with sed-
iments. It should be noted that these factor scores are considered as new uncorrelated variables having
actually the same information content as the original variables. To determine the relationship between
the activities of
226
Ra,
40
K and
137
Cs contained to the second data set and the dominant processes ex-
tracted by the analysis of the first data set, a second-order factor analysis was applied in a new data set
derived from the initial data set (12 12) together with the factor scores of the first order factor analysis
which resulted in a 15 12 matrix. A similar approach was successfully used by Hitchon et al. (1971) to
study the hydrochemical processes of the water formation in a sedimentary basin of Western Canada.
59H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
4. Results and discussion
4.1. Comparison of
238
U and
232
Th activity concentrations measured by INAA
and direct g-ray Spectrometry
The results of
226
Ra,
232
Th,
40
K and
137
Cs activity concentrations measured by direct g-ray
spectrometry (DS) in the sediment samples (n¼12) of Patras Harbour are shown in Table 1.
Table 2 presents the elemental concentrations of uranium and thorium measured by INAA
and the corresponding activity concentrations. It also comprises the measured values by direct
g-ray spectrometry (DS), considering that secular radioactive equilibrium exists among the
nuclides of
238
U and
232
Th series in all sediments samples. The uranium and thorium activities
equivalent to the state of radioactive equilibrium are denoted as e
U
and e
Th
, respectively. The
238
U/e
U
activity ratio is also included in this table.
Comparison between results obtained by the two methods was done by Student’s t-test. The
results showed that the differences in mean values of uranium and thorium, between the two
methods are not statistically significant (P¼0.688; P¼0.156, respectively). As shown in Tabl e 2,
the
238
U/e
U
activity ratio oscillates around one (ratio ranges from 0.83 up to 1.35, mean
238
U/e
U
:
1.02 0.16), indicating that
238
U is close to be in equilibrium with
226
Ra. Student’s t-test also
showed that the differences in mean values of
238
U,
226
Ra and
232
Th between the surface and the
subsurface sediment samples, at the 95% significance level, are not statistically significant
[P¼0.496; P¼0.982; P¼0.052, respectively], while the values of
40
K in subsurface sediments
are significantly higher than those in seafloor surface (P¼0.045).
4.2. Comparison with other areas
Table 3 presents the average specific activities of the natural radionuclides in beach, near-
shore and deep-sea sediments of different countries, as well as the world average (UNSCEAR,
2000). Generally speaking, the radionuclide concentrations in Patras Harbour are lower than
Table 1
Specific activities of
226
Ra,
232
Th,
40
K and
137
Cs in sediment samples of Patras Harbour (units are in
Bq kg
1
d.w. s
tot
)
Sample Depth below the
seafloor (cm)
226
Ra
232
Th
40
K
137
Cs
1/1 5 29.6 4.2 22.5 2.5 452 21 11.5 1.2
1/3 50 22.1 2.0 33.6 1.7 672 25 BD
2/1 5 22.9 1.5 26.8 2.7 516 23 3.9 0.5
2/3 110 25.5 2.6 31.7 1.6 613 24 BD
4/1 5 37.0 3.0 22.9 2.4 434 21 3.2 0.4
4/3 95 23.5 2.1 28.0 2.7 763 25 BD
5/1 5 18.0 2.3 21.7 1.3 426 20 4.8 0.5
5/4 75 22.3 1.0 25.8 1.8 491 20 BD
6/1 5 15.7 1.4 18.0 1.6 387 18 1.9 0.3
6/3 140 22.6 2.0 25.2 1.8 523 22 9.6 1.0
7/1 5 15.5 1.0 15.4 1.5 327 16 2.0 0.3
7/3 50 16.1 1.2 19.3 1.0 363 17 BD
BD: below detection limit.
s
tot
: combined standard uncertain.
60 H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
those measured in three coastal areas of Greece: Cyclades Islands (Florou et al., 1988), Milos
Island (Florou and Kritidis, 1991) and Gulf of Corinth (Papatheodorou et al., 2005). It should be
mentioned that the elevated activities in these islands are due to the presence of volcanic rocks,
while that of central Gulf of Corinth to the submarine bauxitic red mud deposits. On the other
hand, the mean activities in Patras Harbour are comparable to the world average and the activ-
ities from other shallow and deep-water marine environments worldwide (Table 3). These
values could be taken as a baseline data set for Patras Harbour against which future changes
might be compared.
4.3. Areas of erosion, sedimentation and undisturbed seafloor
Patras Harbour consists of three main almost bowl shaped basins (A, B and C) separated by
four docks (D1eD4) (Fig. 1). The detailed bathymetric survey showed that the southern basin
(A), with an extent of 0.16 km
2
and a maximum depth of 10.8 m is the shallowest basin of the
harbour. The central basin (B), extending in an area of 0.14 km
2
with a maximum depth of
15 m, is located between docks D2 and D3 (Fig. 1). The northern basin (C) is the deepest
one with a maximum depth of 16 m (Fig. 1).
Table 2
Comparison of
238
U and
232
Th activity concentrations measured by INAA and direct spectroscopy
Sample Depth below
the seafloor
(cm)
238
U(INAA)
(mg kg
1
)
238
U(INAA)
(Bq kg
1
)
e
U
(DS)
(Bg kg
1
)
232
Th(INAA)
(mg kg
1
)
232
Th(INAA)
(Bq kg
1
)
e
Th
(DS)
(Bg kg
1
)
238
U/e
U
1/1 5 2.0 0.2 24.7 2.5 29.6 4.2 6.1 0.5 24.8 2.0 22.5 2.5 0.83
1/2 40 2.6 0.2 32.1 2.5 ND 6.0 0.5 24.4 2.0 ND e
1/3 50 1.5 0.2 18.5 2.5 22.1 2.0 8.4 0.7 34.1 2.8 33.6 1.7 0.84
1/4 105 1.4 0.1 17.3 1.2 ND 8.4 0.6 34.1 2.4 ND e
2/1 5 2.1 0.2 25.9 2.5 22.9 1.5 7.0 0.5 27.2 2.0 26.8 2.7 1.20
2/2 30 1.7 0.2 21.0 2.2 ND 7.9 0.6 32.1 2.4 ND e
2/3 110 2.3 0.2 28.4 2.5 25.5 2.6 8.0 0.6 32.5 2.4 31.7 1.6 1.11
4/1 5 2.7 0.3 33.3 3.7 37.0 3.0 6.1 0.5 24.8 2.0 22.9 2.4 0.90
4/2 50 1.6 0.1 19.8 1.7 ND 5.7 0.5 23.2 2.0 ND e
4/3 95 2.0 0.2 24.7 2.5 23.5 2.1 7.0 0.6 28.4 2.4 28.0 2.7 0.94
5/1 5 1.3 0.1 16.1 1.2 18.0 2.3 5.8 0.5 23.6 2.0 21.7 1.3 0.82
5/2 25 1.4 0.1 17.3 1.3 ND 5.4 0.4 21.9 1.6 ND e
5/3 35 1.9 0.2 23.5 2.3 ND 4.1 0.3 16.7 1.2 ND e
5/4 75 2.0 0.2 24.7 2.5 22.3 1.0 6.4 0.5 26.0 2.0 25.8 1.8 1.05
6/1 5 1.3 0.1 16.0 1.2 15.7 1.4 4.9 0.4 19.9 1.6 18.0 1.6 1.02
6/2 65 1.2 0.1 14.8 1.1 ND 4.3 0.3 17.5 1.2 ND e
6/3 140 2.0 0.2 24.7 2.5 22.6 2.0 6.3 0.5 25.6 2.0 25.2 1.8 1.09
7/1 5 1.7 0.2 21.0 2.5 15.5 1.0 4.1 0.3 16.6 1.2 15.4 1.5 1.59
7/2 30 1.1 0.1 13.6 1.5 ND 4.2 0.3 17.1 1.2 ND e
7/3 50 1.5 0.2 18.5 2.5 16.1 1.2 4.6 0.3 18.7 1.2 19.3 1.0 1.14
ND: Not determined.
DS: Direct spectroscopy.
INAA: Instrumental Neutron Activation Analysis.
e
U
: Uranium equivalent.
e
Th
: Thorium equivalent.
61H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
Table 3
The mean specific activities of
238
U,
226
Ra,
232
Th and
40
K (in Bq kg
1
) in beach, nearshore and deep-sea sediments for different countries
Country
238
U
226
Ra
232
Th
40
K Sediment type References
Greece (Cyclades Islands) 67 (29e110) 26 (7e159) 31 (4e106) 666 (189e1214) Nearshore sediments Florou et al. (1988)
Greece (Milos Island) 70 (16e119) 75 (19e152) 890 (158e3893) Nearshore sediments Florou and Kritidis (1991)
Greece (Gulf of Corinth) 79.5 (13.1e399.8) 48.7 (12.9e185.2) 85.5 (15.1e412.0) 318.8 (28.3e539.9) Nearshore sediments
and deep-sea sediments
Papatheodorou et al. (2005)
Egypt 56.0 (2.9e61.5) 83.3 (2.3e506.5) 88.1 Beach sand Seddeek et al. (2005)
Egypt 11.7e51.1 11.9e50.6 14.8e67.2 99.4e102.9 Nearshore sediments
(harbour)
Seddeek et al. (2005)
Egypt (Red Sea) 24.7 31.4 427.5 Beach sediments El Mamoney and Khater
(2004)
Sudan (Red Sea) 29.6 (6.5e53.0) 11.6 (2.4e59.9) 6.0 (0.2e19.3) 158.4 (23.7e515.0) Nearshore sediments Sam et al. (1998)
Algeria (Bay of
Ghazaouet)
23.1 74e128 28.6e30.8 446e518 Nearshore sediments Noureddine and Baggoura
(1997)
Bangladesh (Bay
of Bengal)
18e101 138e1318 Nearshore sediments Sharif et al. (1994)
Northwest Pacific Ocean 9.8e43.2 108e1019 13.6e58.6 Deep-sea sediments Moon et al. (2003)
Greece (Gulf of Patras) 21.8 (13.6e33.3) 22.6 (15.5e37.0) 24.5 (16.6e34.1) 497 (327e763) Nearshore sediments This study
World average 35 (16e110) 35 (17e60) 30 (11e64) 1700 (140e850) UNSCEAR (2000)
The results of this study and the international world average are given in the last two rows.
62 H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
This paper does not focus on erosional and accumulation processes; the results, however,
of a bathymetric survey are well suited to interpret the distribution patterns of natural
radionuclides and
137
Cs. Although the changes in harbour depths related to erosion and ac-
cumulation due to ship traffic were not measured, the study of the elaborate bathymetric
map shows explicitly that ship traffic is highly influential as to the harbour bathymetry. Ero-
sion predominates in the deeper parts of the harbour, while deposition and undisturbed sea-
floor have more often been preserved in the shallow parts of the harbour and closer to the
dock sides. The bathymetric survey also found erosional bottoms in rather limited areas
around the terminals (T1eT5). In Patras Harbour, areas of erosion have an elongated pat-
tern which is well correlated to the routes of the ships approaching their terminals (T1eT5)
(Fig. 1). This may be owing to the ship turbulence which is also oriented parallel to the
main direction of the movement. The erosion of the surface sediments increases at the turn-
ing sites (tp) of the passenger ferries close to the docks D3 and D4 and at the centre of the
basin A (Fig. 1). The eroded bottom areas are more extensive in the harbour basin C and
around the terminals T4 and T5. At these sites the very large passenger ferries have their
terminals in the harbour basin C (Fig. 1).
The study of the 3.5 kHz seismic profiles has verified the results of the bathymetric survey
proving the existence of the erosional, depositional and undisturbed areas. The 3.5 kHz profiles
(Fig. 2) show an almost acoustically transparent sequence with a few weak internal reflectors
overlying a highly reflective surface that blocks further seismic penetration. The upper trans-
parent sequence represents Holocene sediments and has a maximum thickness of about
13 m, which gradually decreases towards the dock sides. The transparency acoustic character
of the upper layer indicates a structureless internal stratification and high water content.
The surficial slightly undulated relief, the discontinuous nature and locally the abrupt termi-
nation of the surface reflector suggest that the present surface of the harbour is being eroded by
the turbulence caused by the high-speed ferries (Fig. 1). The underlying acoustic basement rep-
resents the marly bedrock of Patras Harbour. The upper Holocene sequence is less than 5 m
close to the dockside in the central basin and therefore the marly bedrock almost exposures
to the seafloor (Fig. 2).
4.4. Sediment facies
The granulometric and geotechnical parameters of the sediments from Patras Harbour shown
in Table 4. The sediments sampled from the six sediment cores can be divided into two main
groups based on grain size distribution, organic carbon content, colour and geotechnical prop-
erties. The first group consists of sediments from cores 1, 2 and 4 which are located at the
southern (A) and central (B) basins of the Patras Harbour and the second group is composed
of sediments from cores 5, 6 and 7 collected from the northern basin (C) (cores 5 and 6)
and the northern end (core 7) of the harbour.
In cores 1 and 2 from the shallowest southern basin (A) and in core 4 from the central basin
(B) we find the maximum average of organic carbon and water content, while in cores 1 and 2
we find the lower average sand content; in core 4 yet the higher. According to the Casagrande
diagram, the facies found at basins A and B are of dark-gray inorganic mud with medium plas-
ticity to consolidated inorganic to organic mud (BSI, 1975).
The sediments of the second group have a maximum average of sand and a minimum aver-
age of water content and organic carbon. The facies found throughout the whole northern part
of the harbour are inorganic clays with low plasticity to unconsolidated inorganic mud.
63H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
The granulometry of the surface samples (0e5 cm) of recent sediments showed a fining
southward trend in the Patras Harbour (Fig. 3). The presence of coarser and unconsolidated
material at the northern part of the harbour in comparison to that of the southern is likely
due to reworking by the very large passenger ferries. On the other hand, the presence of fine
grained and consolidated sediments dominated in the southern and central basins is in accor-
dance with the low boat traffic.
Vertical profiles of grain size parameters, organic content and geotechnical properties in in-
dividual cores reflect the depositional conditions dominating throughout the harbour. Although
this paper does not deal with heavy metal contamination, the results of a parallel geochemical
survey should be used to help us reach a conclusion on this work (Ferentinos et al., 2003).
Based on that survey, a contaminated surficial layer in the southern and central part of the har-
bour was observed, suggesting that the seafloor is almost undisturbed and/or slightly affected by
the deposition of the new sediments due to the low ship traffic. On the other hand, the absence
of clear organic content and heavy metal concentration downcore trend in the cores from the
northern harbour could be explained by a combination of extensive erosion in the vicinal north-
ern basin, reworking of the sediments in the accumulation areas and possibly new sediment
accumulation. The ship turbulence, especially during turnings (tp) and acceleration at the north-
ern end of the harbour seems to be the principal cause of the extensive erosion and sediment
Fig. 2. 3.5 kHz seismic profiles which show the main features described in the text: (1) sea surface, (2) seafloor surface,
(3) Holocene sequence, (4) Holocene/Pleistocene boundary, (5) seafloor erosional features caused by the turbulence of
the high-speed ferries (see Fig. 1 for location).
64 H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
reworking in the northern basin. The suspended load caused by ship traffic is well distributed at
the surface water due to the turbulence flow. Although, disturbance of the surface water would
facilitate particle flocculation and settling, the high energy conditions because of the ships and
the proximity of the northern basin to the northern opening of the harbour may as well allow the
continuous suspension of the finest fraction and its transportation even out of the harbour. The
northern harbour became more coarse-grained as a result of being covered by the sediments.
Even if deposited, especially the coarsest fraction, the sediment will remain highly vulnerable
to new erosion and reworking until consolidation increases its resistance. This is in accordance
with the unconsolidation of the sediments which have covered the northern harbour. Similar
suggestions have also been reported in the Goteborg Harbour and Gota alv estuary in Sweden
(Brack et al., 2001; Stevens and Ekermo, 2003).
Spatial and temporal variation of gain size characteristics, organic carbon and geotechnical
properties suggest that the erosion and the reworking of the sediments due to ship turbulence
predominate in the northern part of the Patras Harbour. Boat turbulence inside the Patras Har-
bour becomes most decisive in the absence of intense wave agitation and tidal currents.
4.5. Spatial and temporal variation of natural radionuclides and
137
Cs specific activities
Tables 1 and 2 present the specific activities of
238
U,
226
Ra,
232
Th,
40
K, and
137
Cs in Bq kg
1
d.w.
in the sediments of Patras Harbour.
238
U and
226
Ra activity concentrations range from 13.6 to
33.3 Bq kg
1
and from 15.5 to 37 Bq kg
1
with an average value of 21.8 5.5 Bq kg
1
and
22.6 6.2 Bq kg
1
, respectively. The concentrations of
232
Th and
40
K range from 15.4 to
33.6 Bq kg
1
and from 327 to 763 Bq kg
1
with an average value of 24.2 5.4 Bq kg
1
and
Table 4
Granulometric and geotechnical parameters in sediment samples of Patras Harbour
Sample Depth below
the seafloor
(cm)
C
org
(%) Gravel (%) Sand (%) <63 mm
(silt/clay
fraction) (%)
Mean
size (f)
Water
content (%)
Specific
gravity
Bulk
density
(g/cm
3
)
1/1 0 1.42 0.0 6.6 93.4 6.5 54.8 2.75 1.66
1/2 40 2.27 0.3 15.9 83.7 6.2 52.6 ee
1/3 50 1.36 0.0 0.7 99.3 8.0 40.6 2.67 1.73
1/4 105 1.17 0.0 0.5 99.5 7.0 37.2 2.74 2.97
2/1 5 1.49 0.0 8.3 91.6 7.1 63.7 2.78 1.55
2/2 30 1.39 0.0 1.7 98.2 7.5 38.8 2.65 1.81
2/3 110 1.09 0.0 0.4 99.6 7.9 35.5 2.78 3.05
4/1 5 1.29 1.0 20.0 79.0 6.7 45.7 2.82 1.69
4/2 50 1.43 0.0 14.1 85.9 6.4 40.2 2.74 1.74
4/3 95 0.82 0.0 1.1 98.9 6.4 32.8 2.71 1.89
5/1 5 1.26 0.0 23.9 76.1 6.4 eee
5/2 25 1.13 5.3 20.8 73.9 6.2 32.9 2.85 1.69
5/3 35 1.05 0.1 46.5 53.5 5.3 24.6 2.77 1.88
5/4 75 0.88 0.0 19.1 80.9 6.6 33.2 2.76 1.89
6/1 5 1.00 0.0 34.0 66.0 5.8 35.6 2.77 1.74
6/2 65 0.77 2.7 41.9 55.5 5.1 32.7 2.77 1.85
6/3 140 1.08 0.3 13.0 86.7 6.8 eee
7/1 5 e0.2 51.5 48.3 4.9 24.9 2.80 1.90
7/2 30 1.25 0.2 42.9 56.9 5.1 47.3 e1.40
7/3 50 1.40 4.5 46.2 49.2 4.9 34.5 2.70 1.80
f¼log
2
d(cm), d: grain diameter.
65H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
497 130 Bq kg
1
, respectively. The man-made radionuclide
137
Cs was found in measurable
concentrations in all surface sediments (0e5 cm). Although the sediments had been accumulating
in Patras Harbour since the beginning of the atmospheric nuclear weapon testing, none of the fall-
out
137
Cs activities were detected in the subsurface sediments, except one.
137
Cs was detected in
a subsurface sample from core 6, about 1.4 m below the seafloor. The specific activities of
137
Cs in
surface samples range from 1.8 to 11.1 Bq kg
1
with an average of 4.4 3.5 Bq kg
1
.
The spatial distribution of
238
U,
226
Ra,
232
Th,
40
K, and
137
Cs in the surface sediments of Pa-
tras Harbour is shown in Fig. 3. The patterns for
232
Th,
40
K and <63 mm are conformable to
each other indicating an evident ongoing southward trend. At the southern basin were observed
maximum
232
Th and
40
K activities and <63 mm percentage, while the minimum values were
found at the northern basin of the harbour. The latter suggests a direct relationship between
232
Th and
40
K and the fine-grained fraction of the surface sediments. This relationship stipu-
lates explicitly that fine particles, which are mainly consisted of clay minerals, play an impor-
tant role in the distribution of
232
Th and
40
K in the sediments of Patras Harbour. On the other
hand, the areal distribution of
238
U and
226
Ra shows a slightly different pattern. Maximum
238
U
and
226
Ra activities were found at the central harbour basin which is covered by sediments of an
increased sand content (20%). These patterns suggest the occurrence of separate sources for the
two natural radionuclide groups;
238
Ue
226
Ra and
232
The
40
K.
232
Th and
40
K have been
Fig. 3. Aerial distribution of (a) sand percentage, (b) <63 mm percentage, (c)
238
U, (d)
232
Th, (e)
226
Ra, (f)
40
K and
(g)
137
Cs in the surface sediments of the Patras Harbour.
66 H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
introduced in the harbour basins and incorporated mostly in the clay mineral lattice, while
238
U
and
226
Ra have been introduced both by clay minerals and sand fraction.
The
137
Cs distribution in the surface sediments exhibits a clear increasing southward trend,
almost the same with those of <63 mm and C
org
content (Fig. 3). This indicates that clay min-
erals and organic material affect the
137
Cs distribution in the surface sediments. Organic carbon
and clay minerals such as smectite and illite have a large cation exchange capacity (CEC) and,
as a result, would readily absorb cations (Borden and Giese, 2001).
238
U and
232
Th show various changes with depth in sediments, indicating a variety of factors
(granulometry, and mineralogy) and mechanisms (diagenesis) affecting the vertical profiles of
radionuclide concentrations (Fig. 4). In a closed system, like a harbour, with more or less con-
stant mineralogy,
238
U and
232
Th concentrations in sediments should remain fairly invariable
Fig. 4. Vertical profiles of
238
U and
232
Th activities in sediments from the Patras Harbour.
67H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
with depth.
232
Th activities are associated with detrital phases and show only minor changes
with sediment depth indicating small term changes over the time period represented in detrital
phases’ textures and mineralogy. On the other hand,
238
U activities show a greater variability
with sediment depth compared to those of
232
Th, suggesting the increased migration of
238
U
in the sediments and the bottom water of Patras Harbour (Fig. 4). In cores 5 and 6 taken
from the northern harbour,
238
U activities increase with depth, while in cores 1 and 4 from
the southern (A) and central (B) basins,
238
U activities show a clear decreasing downcore trend.
These increases and decreases in
238
U along with the sediment depth are due to the diagenetic
processes in the deeper sediment layers and to the diffusion of the sediment to bottom water
through interstitial water.
238
U activities in sediments of the northern part (C) of the harbour increase with sediment
depth, a fact that suggests that there is an enhanced
238
U addition to deeper layers by diagenetic
processes. Increases are less than those of factor two (w1.5) and are most pronounced in core 5
(Fig. 4). The overlying seawater of an elevated concentration of dissolved
238
U penetrates
several decimetres down into the sediments. These sediment layers become reducing and the
enhanced
238
U deposition in reducing environments is so far well documented (Klinkhammer
and Palmer, 1991). Such ‘‘irrigation’’ of deeper, anoxic sediment layers with oxygenated bot-
tom waters would lead to enhanced
238
U concentrations in deeper sediment layers. This process
has also been reported in the sediments of the Gulf of Thailand and the Southern Baltic Sea
(Srisuksawad et al., 1997; Skwarzec et al., 2004). In the northern basin of the harbour, where
the erosion and the reworking of sediments are more intense due to ship turbulence, the down-
core diffusion of the bottom waters becomes enhanced. In such environment, as suggested by
Srisuksawad et al. (1997), the downcore seawater flow with soluble
238
U due to biological
activity, should be considered negligible.
238
U activities in sediments of the southern basin (A) of the harbour decrease along with
the sediment depth. The surface layers of sediments are penetrated by seawater with soluble
238
U which induces a clear increase of the
238
U activities in comparison with the sediment
layers lying in lower depth. The downcore diffusion of the bottom waters in the southern
basin (A), where the disturbance of the seafloor sediments due to ship turbulence is
negligible, is mainly controlled by biological activity and reaches up to few decimetres
(Fig. 4).
In the vertical profile of
137
Cs of the northern basin (C) of Patras Harbour sediments is not
shown the common downcore decreasing trend (Fig. 4). In this sampling site the higher specific
activity was found in the deeper sediment layers than in the surficial one (Fig. 3). This down-
core trend is in accordance with the high disturbance of the sediments of the northern basin due
to high ship traffic. Due to the fact that the area of a harbour is not undisturbed, no information
on its sediment history could be drawn based on
137
Cs specific activities.
4.6. Multivariate statistical analysis
R-mode factor analysis was applied to the first data set (9 20) of Patras Harbour containing
the
238
U and
232
Th activities, using the log-transformed data. Table 5 displays the varimax
rotated factor loadings for this first data set. Three factors were obtained with eigenvalues sum-
ming up almost to 83% of the total variance.
Factor 1 explains the largest proportion (46.3%) of the total variance and is characterized as
a bipolar factor. The positive pole loads heavily on
238
U and
232
Th, <63 mm percentage, mean
grain size and bulk density. The negative pole represents the variability of the sand fraction in
68 H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
the sediments of Patras Harbour. Factor 1 provides information about the close relationship be-
tween
238
U and
232
Th and the fine fraction of the harbour sediments. Factor 2 which provides
explanations for the 22.3% of the total data variance is also a bipolar factor and loads positively
on organic carbon content and the water content and it is also inversely correlated to the bulk
density. The second factor is considered as an ‘‘organic material’’ factor. Factor 3 accounts for
14.3% of the total variance and has high positive loadings on
238
U and specific gravity. This
loading pattern suggests that the third factor represents the relationship between
238
U activity
and the sediments of high specific gravity. Taking into consideration that this factor shows a low
(0.26) but significant loading on sand fraction, the third factor could be considered as a factor
representing the ‘‘
238
U-bearing coarse-grained sediments’’. This hypothesis is further supported
by the high factor scores obtained in the northern part of the harbour (cores 5 and 6) where
coarse and unconsolidated sediments dominate.
The second-order R-mode factor analysis was carried out using: (a) the factor scores from
the first order factor analysis representing the ‘‘
238
Ue
232
Th-fine sediments’’ association, the
‘‘organic material’’ factor and the ‘‘
238
U-bearing coarse-grained sediments’’ factor and
(b) the second data set consisted not only of
238
U and
232
Th activities, but also of
226
Ra,
40
K
and
137
Cs ones accompanied by the granulometric and geotechnical properties. The choice
of the three most significant factors was done on the basis of a combination of criteria for factor
selection. The three factors explain about 84% of the total variance for log-transformed data
and this percentage is almost equal to that of the first order factor analysis (Table 5). The re-
sulting R-mode varimax factor loading matrix is shown in Table 6.
The first factor accounts for more than 40% of the total variance and shows high positive
loadings for the interpreted ‘‘
238
Ue
232
Th-fine sediments’’ factor and
40
K,
232
Th, mean grain
size, <63 mm, moderate positive loadings for
226
Ra,
238
U and bulk density and negative loading
for the sand fraction. This factor represents the close relationship between the NORM activities
and the fine fraction of the sediments. Most sediment with high clay content is associated with
high specific activities of natural observed radionuclides. Moreover, the close relationship
between natural radionuclides and the bulk density is well documented by many researchers.
Ayers and Theilen (2001) reported a clear trend of increasing bulk density with increasing
natural radionuclides activities and He and Walling (1996) suggested that radionuclides concen-
tration increases as the mean grain size of the sediments decreases. As it is expected, the radio-
nuclide activities are inversely related to the porosity and the water content. Furthermore,
Schuiling et al. (1985) showed that it is in high density materials that appear higher activities.
Table 5
Varimax rotated factor loadings for the first log-transformed data set (first order R-factor analysis)
Variable Factor 1 Factor 2 Factor 3
238
U 0.33 0.24 0.71
232
Th 0.96 0.19 0.04
C
org
0.15 0.83 0.13
Sand 0.91 0.21 0.26
<63 mm 0.90 0.27 0.13
Grain mean size 0.90 0.29 0.16
Water content 0.24 0.85 0.19
Bulk density 0.59 0.61 0.04
Specific gravity 0.30 0.18 0.83
Variance explained by each factor (%) 46.37 22.33 14.26
69H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
The association of radionuclides with fine fraction, mean grain size and bulk density indicates
that the compaction mechanisms which strongly affect the clay sedimentation also affect the
concentration of radionuclides and consequently increase the natural radionuclides activities.
After the deposition and during the compaction caused by the ongoing pressure of the overbur-
den, the trapped water is squeezed out while the radionuclides remain within the clay structure
and/or absorbed onto the clay particles. Therefore, as the water content and porosity decreases
more mass of radioactive material (clay minerals) is concentrated in the same original volume
of the initial unconsolidated material. The distribution of the scores of the first factor supports
further this interpretation. Elevated factor scores are observed at sites 1 and 2 where inorganic
mud of medium plasticity and consolidated inorganic to organic mud are found. Thus, Factor 1
is considered as a ‘‘radionuclides ewell compacted clay’’ factor and is in accordance with the
low ship traffic dominating in the southern harbour basin. The significantly lower factor load-
ings of
238
U and
226
Ra compared to loadings of the
232
Th and
40
K are a result of the different
geochemical behaviour of these radionuclides. The uranium dissolves in water, while thorium is
a particularly insoluble element. The uranium concentration in Patras Harbour sediments
depends on the uranium concentration in the crystal lattice of the minerals and on the mobility
of the uranium derived from the rocks and the marine sediments by rain and seawater, respec-
tively (Strezov et al., 1998; El Mamoney and Rifaat, 2001). The mobility of uranium in the
marine environment is strongly affected by the redox conditions. The oxidised U
6þ
ion
complexes with carbonate, phosphate or sulphate ion and therefore is easily transported. On
the other hand, in reducing waters, U
4þ
has a strong tendency to precipitate and remain immo-
bile (Coward and Burnett, 1994). The well oxidised harbour waters suggest that the uranium
exhibits high mobility in Patras Harbour. Th
4þ
is the only oxidation state of thorium and there-
fore is quite insoluble. The transportation of
232
Th from inland to offshore, as mentioned above,
can occur by the transfer of the particulate matter, to which thorium is incorporated. After the
deposition in the seafloor, the mobility of thorium is restricted. This difference in the geochem-
ical behaviour of
232
Th and
238
U is well established in the sediments of Patras Harbour as
indicated by the loading profile of the ‘‘radionuclides ewell compacted clay’’ factor.
Table 6
Varimax rotated factor loadings for the second log-transformed data set (second-order R-factor analysis)
Variable Factor 1 Factor 2 Factor 3
226
Ra 0.47 0.27 0.69
40
K 0.89 0.21 0.10
137
Cs 0.10 0.52 0.47
238
U 0.35 0.03 0.84
232
Th 0.98 0.01 0.07
C
org
0.06 0.85 0.08
Sand 0.84 0.36 0.19
<63 mm 0.90 0.09 0.28
Grain mean size 0.89 0.10 0.25
Water content 0.12 0.91 0.27
Bulk density 0.40 0.68 0.10
Specific gravity 0.21 0.03 0.92
First order F1 0.96 0.22 0.13
First order F2 0.08 0.99 0.01
First order F3 0.09 0.02 0.98
Variance explained by each factor (%) 40.38 27.81 16.19
70 H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
Furthermore, this interpretation is further supported by the different variability of
232
Th and
238
U with sediment depth (Fig. 4).
The second factor of the second-order factor analysis provides us with a significant portion
(28%) of the total variance of the data and constitutes a bipolar factor (Table 6). Factor 2 has
high positive loadings on the ‘‘organic material’’ factor of the first order analysis, organic
carbon, water content and
137
Cs and high negative loading on bulk density (Table 6). Based
on the loading profile, the second factor could be considered as a ‘‘
137
Cs-binding by organic
particles’’ factor. Ligero et al. (2001) interpreted a similar bipolar factor with a positive pole
of
137
CseC
org
e
40
K and a negative pole of bulk density as a factor representing the granulomet-
ric influence on the element activities and organic carbon.
It is well established that
137
Cs appears to be preferentially bound on clay and organic
particles and is rapidly and strongly absorbed in cation exchange sites (Duran et al., 2004).
A large number of studies have shown that the concentration of
137
Cs in marine sediments
is influenced by grain size, mineral composition and the content of organic matter (Taken-
aka et al., 1998; Hird et al., 1995; Park et al., 2004). In coastal environments,
137
Cs can
deposit on the marine sediments by a wide range of mechanisms including fixation on sus-
pended matter and sedimentation, direct precipitation of colloidal forms, direct binding by
absorption and the deposition of organic wastes. Although
137
Cs is strongly absorbed by the
clay particles which have large surface areas and fine particle sizes and the absorbed
137
Cs
is virtually non-exchangeable, this is not the case of Patras Harbour. The loading pattern of
the second factor suggests that the
137
Cs activity in the sediments of Patras Harbour is
mainly influenced by the organic carbon content. The organic fraction is very substantial
for the binding and fixation of
137
Cs to the sediments. According to Alonso Santos and
Diaz del Rio (1994), the organic matter is important in the process of seabedewater-column
transfer, due to the high number of functional groups of the organic molecules associated
with sediments. The bulk density is negatively correlated with the
137
CseC
org
ewater con-
tent association as indicated by its high negative loading. This suggests that the sediments
with a low bulk density present a higher concentration of
137
Cs and C
org
in Patras Harbour.
Ligero et al. (2001) reported a similar negative correlation between density and
137
CseC
org
association in the seabed sediments of the Bay of Gadiz.
High factor scores are observed at the surface sediments (0e5 cm) of the southern and
central basin of Patras Harbour (cores 1, 2 and 4). The appreciable content (79e93.4%) of
fine particles (<63 mm) in the surface sediments of cores 1, 2 and 4 indicates that fine grains,
especially the clay particles, probably contribute to the observed enrichment of
137
Cs. However,
factor analysis does not detect the relationship between
137
Cs and fine grains due to the specific
vertical profile of
137
Cs in the sediment column, having high activities in the surface sediment
samples and negligible activities in the subsurface sediments.
The third factor accounts for 14% of the total variance and shows high positive loadings
for the interpreted ‘‘
238
U-bearing coarse-grained sediments’’ factor and
226
Ra,
238
U, specific
gravity and moderate positive loading for
137
Cs (Table 6). High positive scores are found at
sediments taken from cores 4, 5 and 6 which are located at the central and northern harbour
basins and are marked by high percentage of sand and high values of specific gravity.
Based on the loading profile and the factor scores, this factor represents the
238
U and
226
Ra containing coarse-grained sediments with high specific gravity. In addition, this sug-
gests that the elevated activities of
238
U and
226
Ra in the sediment cores 4, 5 and 6 (basins
B and C) are possibly due to the heavy mineral grains which are comprised in the sand
fraction of the sediments. Heavy minerals like monazite, apatite, rutile and zircon contain
71H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
high concentration of uranium. The above-mentioned interpretation of the third factor is
fully supported by mineralogical and chemical analyses of the Plio-Pleistocene sediments
of the area. Rathossi (2005) reported the presence of chromite, rutile, apatite and monazite
in the Plio-Pleistocene lacustrineebrackish sediments of the area. Furthermore, chemical
analyses of those sediments at a location of about 1 km eastward of the harbour showed
an elevated concentration of
238
U (mean value: 3.75 mg/kg with SD: 5.54 mg/kg). There-
fore, the third factor could be considered as a factor representing the ‘‘
226
Ra- and
238
U-bearing minerals in coarse-grained sediments with high specific gravity’’. It should
be noted that the Plio-Pleistocene sediments almost exposure at the seafloor in the central
and northern basins (B and C) as indicated in the 3.5 kHz seismic profiles (Fig. 2).
5. Conclusions
The specific activities of
238
U,
232
Th,
226
Ra,
40
K and
137
Cs measured in Patras Harbour
provide us with a representative baseline data regarding the coastal zone of Greece. The rela-
tionship between the activities and the granulometric/geotechnical parameters which explains
the distribution pattern of radionuclides in the sediments of the harbour was studied using
marine remote sensing data and multivariate statistical methods. This multidisciplinary
approach has provided important new insights into sedimentological processes which affected
the distribution of radionuclides in a marine environment highly influenced by human activities.
The granulometric and geotechnical properties of the harbour sediments and consequently the
specific activities of natural radionuclides and
137
Cs seem to be controlled by the ship traffic.
The combination of the results of two separate factor analyses, which were carried out using
log-transformed data, has allowed a meaningful interpretation of the interrelationship between
radioactivity and granulometric/geotechnical properties of the sediments. R-mode factor anal-
ysis has shown that the natural radionuclides and
137
Cs distribution patterns in the harbour
depend on three main sedimentological processes: (i) the association of radionuclides with
fine fraction, mean grain size and bulk density indicating that the compaction mechanisms
which strongly affect the clay sedimentation also affect the concentration of radionuclides
and consequently increase the natural radionuclides activities, (ii) the
137
CseC
org
association
in sediments with high water content and low bulk density and (iii) the
226
Ra- and
238
U-bearing
minerals in coarse-grained sediments with high specific gravity. The areal distribution of these
sedimentological processes is highly affected by the high ship traffic in the harbour.
References
Alonso Santos, J.C., Diaz del Rio, V., 1994. Caracterisation de los sedimentos superficiales en function de su contenido
en material organica: relacion granulometria-materia organica. Cuadernos de Qumica Oceanografica 1, 55e82.
Ayers, A., Theilen, F., 2001. Natural gamma-ray activity compared to geotechnical and environmental characteristics of
near surface marine sediments. Journal of Applied Geophysics 48, 1e10.
Borden, D., Giese, R.F., 2001. Baseline studies of the Clay Minerals Society source clays: cation exchange capacity
measurements by the ammonia-electrode method. Clays and Clay Minerals 49, 444e445.
Brack, K., Johannesson, L.T., Stevens, R.L., 2001. Accumulation rates and mass calculations of Zn and Hg in recent
sediments, Gota alv estuary, Sweden. Environmental Geology 40, 1232e1241.
BSI, 1975. Methods of Test for Soils for Civil Engineering Purposes, BS 1377. British Standards Institution, p. 142.
Christodoulou, D., Papatheodorou, G., Ferentinos, G., Masson, M., 2003. Active seepage in two contrasting pockmark
fields in the Patras and Corinth Gulfs, Greece. Geo-Marine Letters 23, 194e199.
72 H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
Coward, J.B., Burnett, W.C., 1994. The distribution of uranium and thorium decay series radionuclides in the environment e
a review. Journal of Environmental Quality 23, 651e662.
Cundy, A.B., Croudace, I.W., 1995. Physical and chemical associations of radionuclides and trace metals in estuarine
sediments: an example from Poole harbour, Southern England. Journal of Environmental Radioactivity 29 (3),
191e211.
Dunk, R.M., Mills, R.A., Jenkins, W.J., 2002. A re-evaluation of the oceanic uranium budget for the Holocene. Chem-
ical Geology 190, 45e67.
Duran, E.B., Povinec, P.P., Fowler, S.W., Airey, P.L., Hong, G.H., 2004.
137
Cs and
239þ240
Pb levels in the Asia-Pacific
regional seas. Journal of Environmental Radioactivity 76, 139e160.
El Mamoney, M.H., Khater, A.E.M., 2004. Environmental characterization and radio-ecological impacts of non-nuclear
industries on the Red Sea coast. Journal of Environmental Radioactivity 73, 151e168.
El Mamoney, M.H., Rifaat, A.E., 2001. Discrimination of sources of barium in beach sediments, Marsa Alam to
Shuqeir, Red Sea, Egypt. Journal of King Abdulaziz University Marine Sciences 12, 149e160.
Ferentinos, G., Christodoulou, D., Papatheodorou, G., Geraga, M., 2003. Environmental study for the characteristics
and composition of dredging material from Patras Harbour. Technical report submitted to Patras Port Authority
S.A.
Florou, H., Kritidis, P., 1991. Natural radioactivity in environmental samples from an island of volcanic origin (Milos,
Aegean Sea). Marine Pollution Bulletin 22 (8), 417e419.
Florou, H., Kritidis, P., Pavlakis, P., Veldeki, K., 1988. Enhanced natural radioactivity in some areas of the Aegean
Archipelagos. In: Proceedings of the International Conference on Environmental Radioactivity in the Mediterranean
Area, Barcelona, 10e13 May.
Folk, R.L., 1974. Petrology of Sedimentary Rocks. Hemphill’s Publishing Company, Austin, Texas, p. 184.
Hasiotis, T., Papatheodorou, G., Kastanos, N., Ferentinos, G., 1996. A pock-mark field in the Patras Gulf (Greece) and
its activation during the 14/7/1993 seismic event. Marine Geology 130, 333e344.
He, Q., Walling, D.E., 1996. Interpreting particle size effects in the adsorption of
137
Cs and unsupported
210
Pb by
mineral soils and sediments. Journal of Environmental Radioactivity 30, 117e137.
Hird, A.B., Rimmer, D.L., Livens, F.R., 1995. Total caesium-fixing potentials of acid organic soils. Journal of Environ-
mental Radioactivity 26, 103e118.
Hitchon, B., Billings, G.K., Klovan, J.E., 1971. Geochemistry and origin of formation waters in the western Canada sed-
imentary basin eIII. Factors controlling chemical composition. Geochimica et Cosmochimica Acta 35, 567e598.
Klinkhammer, G.P., Palmer, M.R., 1991. Uranium in the oceans: where it goes and why. Geochimica et Cosmochimica
Acta 55, 1799e1806.
Koukis, G., Sabatakakis, N., Tsiambaos, G., Katrivesis, N., 2005. Engineering geological approach to the evaluation of
seismic risk in metropolitan regions: case study of Patras, Greece. Bulletin of Engineering Geology and the
Environment 64, 219e235.
Ligero, R.A., Ramos-Lerate, I., Barrera, M., Casas-Ruiz, M., 2001. Relationships between sea-bed radionuclide
activities and some sedimentological variables. Journal of Environmental Radioactivity 57, 7e19.
Moon, D.-S., Hong, G.-H., Kim, Y.I., Baskaran, M., Chung, C.S., Kim, S.H., Lee, H.-J., Lee, S.-H., Povinec, P.P., 2003.
Accumulation of anthropogenic and natural radionuclides in bottom sediments of the Northwest Pacific Ocean.
Deep-Sea Research Part II: Topical Studies in Oceanography 50, 2649e2673.
Noureddine, A., Baggoura, B., 1997. Plutonium isotopes,
137
Cs,
90
Sr and natural radioactivity in marine sediments from
Ghazaouet (Algeria). Journal of Environmental Radioactivity 34 (2), 127e138.
Papaefthymiou, H., Kritidis, P., Anousis, J., Sarafidou, J., 2005. Comparative assessment of natural radioactivity in
fallout samples from Patras and Megalopolis, Greece. Journal of Environmental Radioactivity 78, 249e265.
Papaefthymiou, H., Symeopoulos, B.D., Soupioni, M., 2007. Neutron activation analysis and natural radioactivity
measurements of lignite and ashes from Megalopolis basin, Greece. Journal of Radioanalytical and Nuclear
Chemistry, in press.
Papatheodorou, G., Hasiotis, T., Ferentinos, G., 1993. Gas charged sediments in the Aegean and Ionian Seas, Greece.
Marine Geology 112, 171e184.
Papatheodorou, G., Lyberis, E., Ferentinos, G., 1999. Use of Factor analysis to study the distribution of metalliferous
bauxitic tailings in the seabed of the Gulf of Corinth, Greece. Natural Resources Research 8 (4), 277e285.
Papatheodorou, G., Hotos, G., Geraga, M., Avramidou, D., Vorinakis, T., 2002. Heavy metal concentrations in
sediments of Klisova lagoon (S.E. MesolonghieAitolikon Lagoon complex), W. Greece. Fresenius Environmental
Bulletin 11 (11), 951e956.
Papatheodorou, G., Papaefthymiou, H., Maratou, A., Ferentinos, G., 2005. Natural radionuclides in bauxitic tailings
(red-mud) in the Gulf of Corinth, Greece. Radioprotection 40, 549e555.
73H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
Papatheodorou, G., Demopoulou, G., Lambrakis, N., 2006. A long-term study of temporal hydrochemical data in a shal-
low lake using multivariate statistical techniques. Ecological Modelling 193, 759e776.
Park, G., Lin, X.J., Kim, W., Kang, H.D., Lee, H.L., Kim, Y., Doh, S.H., Kim, D.S., Yun, S.G., Kim, C.K., 2004. Prop-
erties of
137
Cs in marine sediments off Yangnam, Korea. Journal of Environmental Radioactivity 77, 285e299.
Rathossi, C., 2005. Ancient ceramics from NW Peloponnese and the provenance of their raw materials: a petrographic,
mineralogical, geochemical and archaeometric approach. Unpublished thesis, Department of Geology, University of
Patras.
Reimann, C., Filzmoser, P., 2000. Normal and lognormal data distribution in geochemistry: death of a myth. Conse-
quences for a statistical treatment of geochemical and environmental data. Environmental Geology 39 (9),
1001e1014.
Reyment, R.A., Joreskog, K.G., 1996. Applied Factor Analysis in the Natural Sciences. Cambridge Univercity Press,
Cambridge.
Sam, A.K., Ahamed, M.M.O., El Khangi, F.A., El Nigumi, Y.O., Holm, E., 1998. Radioactivity levels in the Red Sea
coastal environment of Sudan. Marine Pollution Bulletin 36 (1), 19e26.
Schuiling, R.D., de Meijer, R.J., Riezebos, H.J., Scholten, J., 1985. Grain size distribution of different minerals in
sediment as function of their specific density. Geologie en Mijnbouw 64, 199e203.
Seddeek, M.K., Badran, H.M., Sharshar, T., Elnimr, T., 2005. Characteristics, spatial distribution and vertical of gamma-
ray emitting radionuclides in the coastal environment of North Sinai. Journal of Environmental Radioactivity 84 (1),
21e50.
Sharif, A.K.M., Bilkis, A.B., Roy, S., Sikder, M.D.H., Idriss Ali, K.M., Safiullah, S., 1994. Concentrations of radio-
nuclides in coastal sediments from the Bay of Bengal. Science of Total Environment 158, 1e8.
Skwarzec, B., Boryio, A., Struminska, I., 2004. Activity disequilibrium between
234
U and
238
U isotopes in southern
Baltic. Water, Air, and Soil Pollution 159, 165e173.
Srisuksawad, K., Porntepkasemasn, B., Nouchpramool, S., Yamkate, P., Carpenter, R., Peterson, M.L., Hamilton, T.,
1997. Radionuclide activities, geochemistry, and accumulation rates of sediments in the Gulf of Thailand. Continen-
tal Shelf Research 17 (8), 925e965.
Stevens, R.L., Ekermo, S., 2003. Sedimentation and erosion in connection with ship traffic, Goteborg Harbour, Sweden.
Environmental Geology 43, 466e475.
Strezov, A., Milanov, M., Mishev, P., Stoilova, T., 1998. Radionuclide accumulation in near-shore sediments along the
Bulgarian Black Sea coast. Applied Radiation and Isotopes 49 (12), 1721e1728.
Takenaka, C., Onda, Y., Hamajima, Y., 1998. Distribution of
137
Cs in Japanese forest soils: correlation with the contents
of organic carbon. Science of the Total Environment 222, 193e199.
UNSCEAR, 2000. Sources and Effects of Ionizing radiation. United Nations Scientific Committee on the effects of
atomic radiation. Report to General Assembly with Scientific Annexes. United Nations, New York.
74 H. Papaefthymiou et al. / J. Environ. Radioactivity 94 (2007) 55e74
