Impacts of dredged-material disposal on the coastal soft-bottom macrofauna, Saronikos Gulf, Greece

Science of the Total Environment 508 (2015) 320–330

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Impacts of dredged-material disposal on the coastal soft-bottom macrofauna, Saronikos Gulf, Greece N. Katsiaras ⁎, N. Simboura, C. Tsangaris, I. Hatzianestis, A. Pavlidou, V. Kapsimalis Hellenic Centre for Marine Research, Institute of Oceanography, 46.7 km Athens-Sounion Avenue, 19013 Anavyssos, Greece

H I G H L I G H T S • • • • •

Dumping of dredged-material was monitored (prior, during and post). Significant direct (burial) and indirect (smothering) impacts were detected. Degradation remained significant in the spoil-ground, post to dumping. Outside the spoil-ground, the macrofauna diversity indices showed recovery patterns. Elevated contaminants in the area show that benthos remains under stress.

a r t i c l e

i n f o

Article history: Received 27 June 2014 Received in revised form 24 November 2014 Accepted 24 November 2014 Available online xxxx Editor: D. Barcelo Keywords: Dumping Dredged-material Benthos Macrofauna Ecotoxicology Hydrocarbons

a b s t r a c t Dredged sediments derived by the low course and estuary of the metropolitan river of Athens (Kifissos River) were dumped every day for 21 months to an open-sea site in the Saronikos Gulf. The spoil-ground and surrounding area was monitored prior, during and post to dumping for 24 months, over 6-month intervals. Dumping significantly changed the granulometry of the pre-existing superficial sediments to finer-grained only in the spoil ground and increased the sediment contamination load (aliphatic, polycyclic aromatic hydrocarbons and heavy metals) throughout the study area. Microtox® SPT showed that sediment toxicity levels were high at almost all sampling stations. During dumping, burial of natural soft-bottom habitats degraded severely the communities of the spoil-ground resulting in an almost azoic state, as well as significantly declined the species number and abundance of benthic communities in locations up to 3.2 km away from the spoil-ground, due to dispersion of the spoil and smothering. Benthic indices on the surrounding sites were significantly correlated with hydrocarbon concentrations and sediment toxicity levels. Post to dumping, the macrofauna communities of the spoil-ground were still significantly degraded, but the surrounding areas showed patterns of recovery. However, the high concentrations of aliphatic, polycyclic aromatic hydrocarbons and levels of toxicity persisted in the sediments after the ceasing of dumping operations in the study area, implying the ecological hazard imposed on the area. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Dredging and dumping of dredged-material are common human activities causing significant environmental problems in coastal and marine areas (Bolam and Rees, 2003). Several alternate treatments of dredged-material have been examined in order to minimize environmental impacts, such as screening the material clear of contaminants, disposing on naturally unstable habitats, distributing the disposal on shallow layers, or capping contaminated disposed material (for extensive information see: Brannon and Poindexter-Rollings, 1990; UNEP MED/POL, 1999; Essink, 1999; Bolam and Rees, 2003; OSPAR ⁎ Corresponding author at: HCMR, 46.7 km Athens-Sounion Ave., 19013 Anavyssos, Greece. E-mail address: [email protected] (N. Katsiaras).

http://dx.doi.org/10.1016/j.scitotenv.2014.11.085 0048-9697/© 2014 Elsevier B.V. All rights reserved.

Commission, 2008). However, often due to economic considerations, the direct disposal at open sea without treatment is still a priority management option (Harvey et al., 1998). Sediment disposal may be more harmful to benthic communities than any other component of the aquatic ecosystem because of the relative immobility of the benthic organisms (Morton, 1977). Dumping of dredged-material in the marine environment impacts the benthic communities by either directly burying of organisms at the spoil-ground, or indirectly by suspension and relocation of the dredge-spoil (usually described as smothering) in the adjacent areas. Direct burial often leads to immediate mortality, especially when the thickness of the sedimentary cap exceeds 15 cm (Wilber et al., 2007). However, the effective thickness varies according to local species tolerance to sedimentation and the characteristics of the sediment (OSPAR Commission, 2008). When relocation occurs, habitat alterations to a lesser complexity have been

N. Katsiaras et al. / Science of the Total Environment 508 (2015) 320–330

observed due to deposition of fine-grained dredged sediment on coarser-grained natural sediments (Zimmerman et al., 2003). Suspension of the spoil also causes higher turbidity, which may affect growth and survival of benthic organisms in many ways, such as clogging of respiratory and feeding apparatus (Essink, 1999). Furthermore, resuspension of contaminated sediments and remobilization of chemicals can also affect benthic communities (Roberts, 2012). The degree of impact depends on numerous factors, such as the method and intensity of dumping operations, the season, the physical and chemical characteristics of the dredged material, the ecology of species compromising the local communities (reproduction strategy and tolerance to sediment disturbance), and the hydrodynamic and geological state of the spoil-ground (Simonini et al., 2005; references within). Over the last decades, several studies have investigated the impacts of dumping on marine benthic communities (e.g. Harvey et al., 1998; Roberts and Forrest, 1999; Smith and Rule, 2001; Stronkorst et al., 2003; Zimmerman et al., 2003; Witt et al., 2004; Wilber et al., 2007; Bolam et al., 2011); however, due mainly to this variety of governing factors, there are some considerable controversial results among them. In some cases, investigations have demonstrated severe and longtermed changes. For example, Harvey et al. (1998) have detected sediment composition changes and increases of tolerant species at the expense of sensitive species abundance, or Witt et al. (2004) reported a severe decline of species number and the elimination of important habitat structures. On the other hand, other studies have not determined remarkable modifications (e.g. Roberts and Forrest, 1999; Smith and Rule, 2001). Furthermore, the variety of governing factors among different cases of dumping are only part of the reason for controversial results, since it is often overlooked that some studies are focused on one-off placements and recently relinquished sites, while other focus on ongoing disposal activities (Bolam et al., 2011). Among the most sensitive ecological indicators used to detect impacts by dumping are the species number (S), the abundance (N) and Shannon–Wiener diversity (H′); their high linkage to such human activities and physical disturbances was also verified by a performance evaluation of eleven different indicators (Ware et al., 2008). Other forms of recorded impacts are shifts in dominance patterns between sensitive and opportunistic species, feeding traits and the elimination of important habitat structures (Harvey et al., 1998; Witt et al., 2004; Simonini et al., 2005; Simboura et al., 2007). In addition to benthic community structure, geochemical analyses and toxicity bioassays can be carried out to determine dumping effects of contaminated dredged-material (Chapman, 1990), although contaminant concentrations are not always correlated with macrofauna responses (e.g. Roberts and Forrest, 1999; Stronkorst et al., 2003; Bolam et al., 2011). The divergent results obtained in the various study areas show that the potential environmental effects must be evaluated on a case-by-case base (Harvey et al., 1998). In addition, most of the above studies occurred in North Atlantic and Baltic Sea, while the number of articles referring to the effects of dumping dredged-material to macrofauna of Mediterranean habitats is considerably limited (e.g. Toumasiz, 1995; Simonini et al., 2005). The present paper aims to describe the benthic community response to dredged-material disposal, by estimating the degree of impact on macrofauna derived by the physical and chemical pressure of discharged sediments and assessing recovery patterns in the casestudy of Saronikos Gulf. 2. Material and methods 2.1. Study area and sampling design Inner Saronikos Gulf surrounding Athens metropolitan area is subject to a number of anthropogenic pressures, with urban waste effluents and resulting organic enrichment being the main source of pollution (Simboura et al., 2014). The degradation of benthic communities due to urban wastes around the study area is well studied and opportunistic

321

species range within 62–75% of the total macrofauna abundance (Simboura et al., 2005, 2014). Nearby, Kifissos River operates as the major drainage channel for the city of Athens. The river bed and flow are completely altered by artificial embankments and other constructions. A recent evaluation of sediment quality in the river lower course and estuary demonstrates relatively elevated aliphatic and polycyclic aromatic hydrocarbon concentrations, but low heavy metal concentrations (except of Cu and Zn) (Panagiotopoulos et al., 2010). In the same study, grain-size distribution has showed that fine-grained sediments (muddy sand, sandy mud and mud) dominate in the area around the river mouth. From May 2010 to January 2012, dredging of the Kifissos estuary had as a result the production of significant amounts of sediment, which were licensed to be dumped further seawards in a designated open sea area of surface 1 nmi2 (ABCD area, Fig. 1). For 21 months, a total of ~ 700,000 m3 of dredged-material were dumped, with a mean monthly discharge of 33,333 m3. The two factors taken into account in the sampling design are the distance from the spoil ground (inside and outside the spoil ground) and time in relation to dumping operations. Lack of a possible control site (according to Before-After-Control-Impact approach; Underwood, 1991) is due to the absence of an undisturbed similar biotope in terms of depth and substrate in the area. Saronikos Gulf is characterized by a deep western sector with homogenous muddy substrates and eastern heterogeneous substrates with lower depths (Simboura et al., 2014). In addition, Saronikos Gulf is influenced by several anthropogenic coastal activities which should be considered when selecting a control site. In total, five stations were sampled (Fig. 1). Two stations (Stations 1 and 2) were located inside the licensed disposal area and the other three stations (Stations 3, 4 and 5) were located in the surrounding area, outside the spoil-ground. Positions and depths of sampling station are given in Table 1. The sampling prior of dumping was done in April 2010. During dumping the stations were monitored over 6-month intervals (October 2010, April 2011, October 2011), while a sampling was done 4 months after dumping operations ceased (April 2012). Although dumping operators were licensed for the whole ABCD area, until October 2011 only the location of Station 1 was used. 2.2. Macrofauna Two replicate samples were collected from each station with the use of a Van Veen grab (0.1 m2) and were washed separately through a 1.0 mm sieve. Residuals were preserved with a buffered 4% formalin/ seawater solution, stained with Rose Bengal. At the laboratory, organisms were sorted from the sediment and were identified to species level. 2.3. Environmental variables Water samples were collected from the bottom-layer of the water column, using Niskin bottles of 8lt. Measurements of dissolved oxygen (D.O.) were performed immediately after the sampling using the Winkler method modified by Carpenter (1965). The quality control/ quality assurance (QC/QA) was achieved with the daily standardization of thiosulphate solution with ‘fresh’ standard solution of potassium iodide. The precision of the method is estimated at 2.2 μmol O2/L. Sediment samples were collected from the uppermost 2 cm of the sea bottom using a Van Veen sampler. The granulometric analysis (applying wet sieving) of the samples comprised of separation of the coarse-grained fraction (N 63 μm) from the fine-grained fraction (b 63 μm). Further classification of the sand and mud fractions was accomplished by the use of a standard set of sieves and a grain-size analyzer (Sedigraph 5100), respectively. At silt abundance N 5%, the precision and accuracy of both weight percentage and mean grain size are estimated at b 5% (Bianchi et al., 1999). Sediment texture was classified according to Folk (1974) nomenclature. Polycyclic aromatic (PAH) and aliphatic hydrocarbon (AH) concentrations were measured by gas chromatography–mass spectrometry on an Agilent 7890 GC, equipped

322

N. Katsiaras et al. / Science of the Total Environment 508 (2015) 320–330

Fig. 1. The study area (Saronikos Gulf) with the location of the sampling stations 1–5 (black dots). The box ABCD designates the licensed area for disposal of dredged-material.

with an HP-5MS capillary column (30 m × 0.25 mm i.d. × 5 0.25 μm phase film) coupled to an Agilent 5975C MSD as described in Kapsimalis et al. (2010) and Botsou and Hatzianestis (2012). The accuracy of the PAH determination was evaluated by analyzing a National Institute of Standards (NIST) standard reference sediment SRM 1941a (Organics in Marine Sediment). The determined values ranged between 94.4% and 107.5% of the certified values. The precision, evaluated in terms of repeatability of the experimental results (n = 7) for the analysis of a real sample and expressed in terms of relative standard deviation, was 5.8% for the UCM and total resolved compounds and ranged between 1.7 and 6.3% for the PAhs. Heavy metal concentrations were determined by X-Ray fluorescence analysis (XRF) using a Philips PW2400 XRF system, with the relative uncertainties of the method being within 5.0% (Karageorgis et al., 2005). 2.4. Sediment toxicity The Microtox® solid phase test was applied for whole sediment toxicity testing according to the standardized SPT protocol (Azur

Environmental, 1998). Freeze-dried bacteria (Vibrio fischeri) and all materials and reagents were purchased from SDIX (USA). Briefly, 7 g of sediment was mixed with 35 ml of solid-phase diluent, stirred for 10 min, and used to make up sediment dilutions in borosilicate tubes. Bacteria were ex-posed to each one of the sediment dilutions for 20 min at 15 °C. At the end of the exposure period, a column filter was inserted in the tubes in order to separate the liquid phase from the sediment and the light output of the liquid phase containing the exposed bacteria and it was measured in a Microtox® analyzer (Azur Environmental, USA). Each sediment dilution was tested in duplicate. A basic test was conducted with the reference standard ZnSO4 to ensure the validity of the test method. The Microtox Omni Software was used to calculate EC50 expressed as concentration (mg l−1) of whole sediment corrected for moisture content, which causes a 50% reduction in bioluminescence. Sediments showing EC50 values below the limit of 1000 mg l−1 previously set for sediment toxicity for the Microtox® SPT test (Bombardier and Bermingham, 1999; Environment Canada, 2002) were considered toxic. 2.5. Statistical analysis

Table 1 Depth and coordinates of the sampling stations.

Depth (m) Latitude Longtitude

Station 1

Station 2

Station 3

Station 4

Station 5

66 37°53′40″N 23°39′40″E

69 37°53′10″N 23°39′10″E

50 37°53′20″N 23°40′50″E

77 37°52′20″N 23°38′20″E

71 37°54′00″N 23°38′20″E

Based on the results of the macrofauna analysis, the following indices were calculated: number of species (S), abundance (N), Pielou's evenness (J′), Shannon–Wiener Diversity (H′) and biotic index Bentix, developed by Simboura and Zenetos (2002). No transformations were applied. The categorizing of species to feeding guilds was done mainly according to Fauchald and Jumars (1979). All diversity indices were

N. Katsiaras et al. / Science of the Total Environment 508 (2015) 320–330

calculated by using Primer 6 and Bentix by using the MS Office add-on (available at www.hcmr.gr). Bentix is a biotic index based on the ecological model by Grall and Glémarec (1997), classifying the benthic macrofauna into ecological groups according mainly to their sensitivity to organic enrichment, but also to other common anthropogenic disturbances. It takes into account two general groups of taxa, GS (Group Sensitive) and GT (Group Tolerant), with their proportional abundance percentage according to the following formula: (6 × %GS + 2 × %GT) / 100. It was developed and intercalibrated (GIG, 2013) for the assessment of the ecological quality status under the Water Framework Directive (EC, 2000). Values from 6 to 4.5 correspond to “High”, from 4.5 to 3.5 to “Good”, from 3.5 to 2.5 to “Moderate”, from 2.5 to 2 to “Poor” and finally zero values to “Bad” ecological quality status. The software of Bentix gives an acceptable confidence level on a matrix with at least 3 species and 6 specimens but a matrix containing at least 10 species is recommended for a higher level of confidence. To examine the dumping effects in the study area, the 6-month sampling intervals were grouped into three sampling periods: prior, during and post to dumping. Firstly, significant alterations were generally investigated in biological and environmental variables between sampling periods and stations. Statistical significance was set to p b 0.05. In detail, the normality of all macrofauna indices data, EC(50) values and environmental variables (Bottom Layer Dissolved Oxygen, sediment concentrations of Cu, Zn, AH and PAH) were tested by Shapiro–Wilk test and transformations were applied where needed. Homogeneity of variance was tested using Levene's test. An initial one-way ANOVA was applied using biological and environmental parameters as dependant variables and sampling periods as independent variable. In the few cases where normality and homogeneity could not be succeeded after transformation, Kruskal–Wallis test was applied instead. Post-hoc tests of Fisher's LSD or Dunnett's T3 were used according to variance homogeneity test results. The parameters showing a significant difference were used in a second set of one-way ANOVA, or Kruskal–Wallis tests, using stations 1–5 numbers as independent variable and post hoc tests (Fisher's LSD or Dunnet's T3) were also applied. Clusters between stations were also found. Bray–Curtis similarity (group average) was applied to the macrofauna species abundance matrix of all stations (prior, during and post to dumping), after the data were transformed by log(x + 1). Cluster dendrogram and MDS plots were constructed. Primer 6 was used for Bray–Curtis similarity, Cluster dendrograms and MDS plots, while Statgraphics Centurion 2009 software package (Statpoint Technologies, Inc.) was used for Levene, Shapiro–Wilk, ANOVA, Kruskal–Wallis and post-hoc tests. Results of spatial differences between stations were used to group the stations according to different types of impact caused by dumping (direct impact of burying inside the spoil-ground and indirect impact of spoil suspension and smothering outside the spoil-ground). Severe mortality caused by physical burying can confound the impact and recovery patterns. Therefore, the assumption that the group of stations outside the spoil ground show different significance of impacts, contaminant concentrations and recovery patterns were tested. All the above analyses were re-applied for data only from stations outside the spoilground (influenced by spoil suspension and smothering). In addition, correlations between biological and environmental variables were examined for stations outside the spoil-ground, using two-tailed Spearman Rank Correlation. Data from inside the spoil-ground were excluded, since mortality due to physical burying could confound the correlation with increases of contaminant concentrations. 3. Results 3.1. Environmental variables All environmental variables are listed in Table 2. Table 3 presents the results of the one-way ANOVA or Kruskal–Wallis and post-hoc tests of

323

biological and environmental data with sampling periods and stations treated as independent variables. A significant change (p b 0.001) of D.O. content was recorded during dumping, but no significant differences were found between stations. D.O. showed a delayed significant change (period pairs during/post) and remained different after cease of dumping operations (prior/post). Prior to dumping, the surface sediments of the study area were dominated by the coarse-grained fraction (ranging from 70% to 87%). Dumping operations caused significant alterations (p b 0.05), which were significantly different (p b 0.0001) between stations inside ABCD area (Stations 1 and 2) and the outer stations according to post-hoc tests. The fine-grained fractions (b 63 μm) increased in Station 1 from 30 to 57.9%, while in Station 2, remained almost constant ranging from 28 to 30.7%. However, post to dumping at Station 2 the fine-grained material raised up to 52% (Table 2). For sediment contaminants, significant increases of AH (p b 0.001) and PAH (p b 0.001) were found after the start of dumping operations, which remained significantly different than initial state after the cease of operations, as it shown in post-hoc test results (period pairs of prior/during and prior/post). In spatial scale, significant differences of AH (p b 0.01) and PAH (p b 0.001) were found only between concentrations of Station 1 and the rest of the stations, as it is evident by post-hoc tests results (Table 2). In Station 1, AH concentrations increased from 94.8 to 1853.2 μg g− 1 (p b 0.05) and PAH concentrations increased from 451.7 to 6175.4 ng g−1 (p b 0.05) during dumping. Post to dumping, AH and PAH concentrations were found to be 1003 μg g−1 and 3394 ng g−1 respectively. 3.2. Sediment toxicity Microtox® SPT EC(50) values in sediments per station and sampling periods are shown in Table 2. The lowest EC(50) values of the Microtox® SPT test during and post to dumping, indicating the highest toxicity levels, were recorded in the spoil-ground (Station 1), ranging from 167 to 604 mg l−1. However, low EC(50) values below the limit of 1000 mg/l set for sediment toxicity for the Microtox® SPT test (Bombardier and Bermingham, 1999; Environment Canada, 2002) were recorded in Stations 2, 4 and 5 during and post to dumping, with the exception of a value just over the limit (1054 mg l−1) in Station 2. Station 3 showed the highest EC(50) values, ranging from 981 to 1819 mg/l. ANOVA (Table 3) showed that there were no significant differences of toxicity levels between sampling periods for all stations, but no data exist for prior to dumping. 3.3. Macrofauna A total of 2490 individuals belonging to 220 taxa were identified. Polychaetes were the most abundant group (85.55% of total abundance), followed by Crustacean (6.84%), Molluscs (4.5%) and other phyla (3.11%). This ranking was similar to all stations. The species with the maximum recorded abundance per station were the polychaetes Paralacydonia paradoxa (39 individuals), Monticellina dorsobranchialis (19 ind.), Levinsenia gracilis (19 ind.), Chaetozone gibber (17 ind.), Aphelochaeta marioni (11 ind.), Paraprionospio coora (9 ind.) and Lysidice unicornis (7 ind.) which are opportunistic or tolerant to disturbance species and Aponuphis brementi (19 ind.), which is a sensitive species (Simboura and Nicolaidou, 2001; Simboura and Zenetos, 2002). The highest diversity and abundance were recorded in Station 1 (60 ± 3 species, 239 ± 66 individuals) prior of dumping, where the community turned to almost azoic conditions (minimum of 1 ± 0 species, 1 ± 0 individuals), during dumping operations. Few species and individuals were found in Station 1 even when dumping operations ceased. Among the other stations, Stations 2 and 5 exhibited the lower diversity and abundance post to dumping (Table 2). ANOVA revealed that the dumping operations caused significant differences in species numbers (S, p b 0.001), abundance (N, p b 0.05) and

324

N. Katsiaras et al. / Science of the Total Environment 508 (2015) 320–330

Table 2 Values of environmental (Cu, Zn, AH, PAH, D.O., Mud), values of toxicity (EC50) with 95% confidence level and mean values of biological (S, N, J′, H′, Bentix) parameters with standard deviation per sampling period, sampling interval and station. Sampling periods

Environmental variables Stations

Prior During

Post Prior During

Post Prior During

Post Prior During

Post Prior During

Post

Biological variables

Cu

Zn −1

μg g

AH −1

PAH −1

(Sampling intervals)

μg g

Station 1 (Apr 10) Station 1 (Oct 10) Station 1 (Apr 11) Station 1 (Oct 11) Station 1 (Apr 12) Station 2 (Apr 10) Station 2 (Oct 10) Station 2 (Apr 11) Station 2 (Oct 11) Station 2 (Apr 12) Station 3 (Apr 10) Station 3 (Oct 10)

12 49 185.2 53 58 14 28 25.9 11 19 7 6

59 200 188.3 261 200 58 132 93.5 11 54 27 28

94.8 1287 895.3 1853 1003 102.5 420 325.6 87.3 754.1 54.5 121.1

Station 3 (Apr 11)

11.5

31.3

Station 3 (Oct 11) Station 3 (Apr 12)

6 9

Station 4 (Apr 10) Station 4 (Oct 10) Station 4 (Apr 11) Station 4 (Oct 11) Station 4 (Apr 12) Station 5 (Apr 10) Station 5 (Oct 10) Station 5 (Apr 11) Station 5 (Oct 11) Station 5 (Apr 12)

6 13 10.5 23 11 17 15 16.3 11 19

mg g

ng g

D.O. −1

Mud EC(50)

S

−1

mg/L %

mg l

451.7 5549.8 4302 6175.4 3394 614 1965 1406 361.5 1881 159.2 195.2

7.6 7.1 7.4 7.1 7.6 7.3 6.7 7.0 7.4 8.0 7.9 6.9

30 52.1 57.9 47.3 16.5 28 30.7 28.9 28 51.7 15 15.6

80.9

453.5

7.7

20.9

19 29

111.9 124.8

548.3 182.6

7.0 8.0

18.2 17

31 51 33.5 34 35 59 65 55 37 53

48.9 164.5 94.1 131.3 345.1 51.7 189 130.9 176.2 702.3

260.1 733.3 351.1 471 470 412.5 984.8 2293.7 1055.2 1492.1

6.4 6.7 7.9 7.3 8.1 7.4 6.4 6.7 6.3 7.9

13 24.7 18 16.1 15 24 26.6 21.4 20.2 21.3

– 167 (95–290) 259 (208–327) 604 (584–625) 396 (360–426) – 797 (360–1766) 448 (366–549) 1034 (699–1530) 537 (413–699) – 1819 (1487–2225) 1254 (1046–1508) 981 (792–1215) 1101 (1031–1802) – 752 (663–853) 175 (100–289) 530 (448–627) 542 (470–626) – – 348 (279–433) 499 (263–409) 528 (373–747)

N

J′

0.1 m

2

0.1 m

60 1 4 8 17 41 26 20 43 36 50 27

3 0 1 3 5 14 16 3 8 1 3 1

239 1 5 11 30 106 58 28 119 81 138 50

± ± ± ± ± ± ± ± ± ± ± ±

45 ± 7

2

± ± ± ± ± ± ± ± ± ± ± ±

0.1 m 66 0 1 7 10 37 44 1 19 6 2 15

± ± ± ± ± ± ± ± ± ±

0.86 0.00 0.98 0.94 0.88 0.91 0.90 0.96 0.89 0.90 0.81 0.90

± ± ± ± ± ± ± ± ± ± ± ±

Bentix

0.1 m 0.01 0.00 0.02 0.09 0.10 0.01 0.00 0.00 0.05 0.02 0.03 0.05

5.06 0.00 2.13 2.81 3.62 4.83 4.12 4.20 4.84 4.65 4.83 4.33

2

± ± ± ± ± ± ± ± ± ± ± ±

0.1 m2 0.01 0.00 0.18 0.3 0.76 0.51 0.85 0.25 0.53 0.18 0.12 0.29

3.25 2.00 5.00 3.50 3.30 3.17 3.41 3.15 3.18 3.16 3.11 3.47

± ± ± ± ± ± ± ± ± ± ± ±

0.11 0.00 1.41 0.71 0.61 0.02 0.30 0.26 0.14 0.13 0.24 0.24

133 ± 16 0.83 ± 0.03 4.58 ± 0.34 3.23 ± 0.49

34 ± 10 75 ± 6 46 ± 6 114 ± 3 34 31 18 31 41 37 37 37 28 30

H′ 2

0.91 ± 0.02 4.6 ± 0.5 3.45 ± 0.15 0.86 ± 0.09 4.76 ± 0.64 3.59 ± 0.12

1 70 ± 7 0.86 ± 0.05 8 74 ± 42 0.90 ± 0.06 8 40 ± 6 0.87 ± 0.11 6 86 ± 8 0.80 ± 0.04 17 111 ± 31 0.86 ± 0.01 3 161 ± 14 0.88 ± 0.03 18 91 ± 62 0.91 ± 0.03 2 106 ± 24 0.84 ± 0.02 1 78 ± 21 0.87 ± 0.04 1 71 ± 1 0.85 ± 0.02

4.38 4.44 3.63 3.98 4.54 5.08 4.68 4.39 4.22 4.13

± ± ± ± ± ± ± ± ± ±

0.25 0.04 0.98 0.44 0.46 0.24 0.49 0.01 0.15 0.08

3.31 3.14 2.69 3.12 3.01 2.93 3.63 2.69 2.97 2.92

± ± ± ± ± ± ± ± ± ±

0.50 0.31 0.19 0.11 0.03 0.01 0.59 0.08 0.15 0.11

Table 3 Tests of homogeneity, variance and post-hoc tests of environmental and biological variables with sampling periods and stations as factors for all stations. Significant differences are indicated with bold letters. (df): degrees of freedom, (F): F-test ratio, and (p): p-value. Sampl. per.: Sampling Periods. Levene's Factors

Test

ANOVA p

df

F

p

S

Sampl. Per. Stations

0.52 2.33

0.5973 0.070

49 49

9.9 9.9

0.0003 0.013

N

Sampl. Per. Stations

2.42 3.74

0.1 0.010

49

8.2

0.0009

J

Sampl. Per.

1.30

0.28

47

1.59

0.2149

H

Sampl. Per. Stations Sampl. Per. Sampl. Per. Sampl. Per. Sampl. Per.

0.07 2.40 2.55 5.18 0.59 5.82

0.9359 0.065 0.088 0.028 0.551 0.005

46 46 49

4.97 1.703 0.88

0.0113 0.167 0.4232

37

0.04

0.8361

AH

Sampl. Per. Stations

6.68 6.09

PAH

Sampl. Per. Stations

DO

Mud%

Bentix EC(50) Cu Zn

Kruskal–Wallis

Post-hoc test periods

Statistic

Periods pairs

Difference

Prior–during Prior–post During–post

21.700 14.000 −7.700

Prior–during Prior–post During–post

4.382 2.866 −1.516

Prior–during Prior–post During–post

200.696 154.231 −46.465

10.2828

p

0.036

0.01761

0.894

0.241416

0.8863

0.003 0.0005

21.8716 17.142

0.0001 0.002

Prior–during Prior–post During–post

−277.867 −685.420 −407.553

7.78 11.11

0.001 0.0002

9.68944 26.370

0.0079 0.0001

Prior–during Prior–post During–post

−1410.227 −1104.400 305.827

Sampl. Per. Stations

2.84 4.57

0.068 0.003

21.884 4.961

0.0001 0.291

Sampl. Per. Stations

0.49 6.59

0.616 0.0002

Prior–during Prior–post During–post Prior–during Prior–post During–post

0.280 −0.600 −0.880 0.02255 −0.01445 −0.03701

49

3.25

0.048 27.6912

0.0001

Post-hoc test stations 1

3

4

5

2 3 4 5 2 3 4 5

−15.000 −22.200 −13.100 −19.100 −3.099 −4.491 −3.111 −4.427

−7.200

1.900 9.100

−4.100 3.100 −6.000

−1.391

−0.012 1.380

−1.328 0.063 −1.317

2 3 4 5 2 3 4 5

688.780 926.600 869.940 776.660 2729.060 3666.800 3517.460 2726.900

237.820

181.160 −56.660

87.880 −149.940 −93.280

937.740

788.400 −149.340

−2.160 −939.900 −790.560

2 3 4 5

−0.00786 −0.07247 −0.06092 −0.07269

−0.06461

−0.05307 0.01155

−0.06483 −0.00022 −0.01177

N. Katsiaras et al. / Science of the Total Environment 508 (2015) 320–330

Shannon–Wiener diversity (H′, p b 0.05). For species number and abundance, post-hoc tests showed that they remained significant different than the initial state, after the cease of operations (sampling period pairs prior/during and prior/post). In the spatial scale, significant differences of species number (p b 0.05) and abundance (p b 0.05) were found and were explained by post-hoc tests because of the differences between Station 1 and the rest of the stations (Table 3). Cluster dendrogram and MDS plot (Fig. 2) show that Stations 2, 3, 4 and 5 of all sampling periods and Station 1 prior to dumping, were clustered together with a similarity of over 40%. Station 1, in the spoilground, during and post to dumping seem to differentiate in relation to the rest of the stations (similarity lower than 30%) and to sampling periods. The decrease of abundance was accompanied by a decline in both opportunistic and sensitive species at all stations. Non-significant variations of the order of ±8% in all stations were observed. Bentix did not show a clear pattern in relation to the disposal of dredged-material, classifying the majority of sampling sites in “Moderate” Ecological Status prior (values range from 2.5 to 3.5), during and post to dumping. The feeding traits of species were characterized by dominance of deposit feeders (surface and subsurface) for all stations (ranging from 51% to 76.8%). Neither significant shifts in this dominance pattern, nor a significant decline of filter-feeders was observed, except at the spoil ground (Station 1), where a few species (mostly carnivores of high motility) were recorded during dumping. 3.4. Impacts outside the spoil-ground by re-suspension and smothering The cluster of Stations 2, 3, 4 and 5 was used to test the assumption that impacts by re-suspension of the spoil and smothering show different significance of impacts, contaminant concentrations and recovery

325

patterns. ANOVA result testing differences caused by assumed spoil re-suspension and smothering, among periods and stations are shown in Table 4. Significant differences were found for species number (p b 0.05) and abundance (p b 0.05), between sampling periods. Excluding the spoilground (Station 1), different patterns of recovery are shown (Table 4). Post-hoc tests show indications of recovery, since no significant differences were found between prior and post to dumping sampling periods (Table 4, Fig. 3). Based on the benthic indices ranges, the most affected station outside the spoil-ground seems to be Station 5, where species number (S) and abundance (N) ranged from 37 ± 3 and 161 ± 14 respectively prior to dumping, to 28 ± 1 species and 71 ± 1 individuals during dumping. However, ANOVA testing spatial scale showed no significant differences among stations (Table 3). Regarding environmental parameters outside the spoil-ground, significant differences were found in concentrations of AH (p b 0.0001), PAH (p b 0.05) and D.O. (p b 0.001) over sampling periods (Table 4). Hydrocarbons (AH and PAH) significantly changed during dumping and remained different after cease of dumping operations (significant period pairs prior/during, prior/post). D.O. showed a delayed significant change and remained different after cease of dumping operations (significant period pairs prior/post, during/post). Significant spatial differences (Table 4) were found only for AH (p b 0.001) and PAH (p b 0.0001). For PAH, post-hoc tests showed no significant differences between Station 2 and Station 5, but both were found significantly different towards all the other pairs (Table 4). Generally outside the spoilground, the more elevated concentrations were found at Stations 2 and 5, reaching maximums of 1881 μg g−1 and 2293.7 μg g−1 respectively (Table 2). Bray–Curtis similarity revealed that the macrofauna communities from prior and post to dumping samplings tend to fall close to each

Fig. 2. Cluster dendrogram and MDS plot of all stations and sampling periods. Pr: prior-dumping, D: during-dumping, Po: post-dumping.

326

N. Katsiaras et al. / Science of the Total Environment 508 (2015) 320–330

Table 4 Tests of homogeneity, variance and post-hoc tests of all environmental and biological variables with sampling periods and stations as factors, outside the spoil-ground (Stations 2, 3, 4, 5). Significant differences are indicated with bold letters. (df): degrees of freedom, (F): F-test ratio, and (p): p-value. Samp. Per.: Sampling Periods. Levene's

ANOVA

Factors

Test

p

df

F

p

S

Sampl. Per. Stations

0.092 0.521

0.9126 0.597

37 2

5.21 1.62

0.0104 0.2042

N

Sampl. Per. Stations Sampl. Per. Sampl. Per. Sampl. Per. Sampl. Per. Sampl. Per. Sampl. Per.

0.976 1.380 1.017 0.711 0.725 0.797 0.831 2.113

0.3867 0.261 0.3719 0.4979 0.4911 0.3802 0.4419 0.1359

37 2 37 37 37 27

3.99 1.49 0.91 2.23 0.06 0.01

0.0274 0.2364 0.4127 0.1228 0.9384 0.9038

37

0.04

0.9602

AH

Sampl. Per. Stations

0.435 0.606

0.6503 0.550

37 3

23.89 7.75

0.00001 0.0005

PAH

Sampl. Per. Stations

0.972 3.729

0.3881 0.031

37

3.5

0.041

Sampl. Per. Stations Sampl. Per.

3.517 2.843 5.450

0.0406 0.068 0.0008

J H Bentix EC(50) Cu Zn

DO Mud%

Kruskal–Wallis

Post-hoc test periods

Post-hoc test stations

Statistic

Periods pairs

Difference

2

4

5

Prior–during Prior–post During–post Prior–during Prior–post During–post

13.13 6.04 −7.08 40.54 19.82 −20.68

Prior–during Prior–post During–post Prior–during Prior–post During–post Prior–during Prior–post During–post Prior–post

0.045 0.069 0.024 −0.783757 −0.454292 0.329465 0.25 −0.75 −1.00 −0.01445

−0.0344 −0.0237 −0.01313 1.19243 0.72204 −0.19691

0.010686

0.02127 0.010587

−0.4704

−1.3893 −0.9189

1.17123

p

0.5568

17.786

0.0005

14.8864 4.513 4.845

0.0006 0.211 0.0886

other and separate from most communities during dumping (Fig. 4). However, Station 5 clustered together with Station 3 prior to dumping, but during and post to dumping was found to cluster with Station 4, showing similarity of over 50% in all cases. The clustering of stations in the MDS plots corresponds to projected Microtox® SPT EC(50) values, reflecting community clusters with similar toxicity levels. Station 5 communities were modified and separated from Station 3 towards Station 4 as local Microtox® SPT EC(50) values decreased over time (Table 2).

3.4. Biotic–abiotic interactions Spearman Rank Correlation showed negative significant correlations of species number (S), abundance (N) and Shannon–Wiener diversity (H′) with AH (p b 0.05, p b 0.05 and p b 0.05, respectively). There were also significant positive correlations of species number (S), Shannon diversity (H) and Bentix with Microtox® SPT EC(50) (p b 0.05, p b 0.05 and p b 0.001, respectively). Also, Microtox® SPT EC(50) showed significant negative correlations with Cu, Zn, AH and PAH (p b 0.05, p b 0.01, p b 0.05, p b 0.01 respectively). Regarding

3 4 5 3 4 5

water column variables, bottom layer dissolved oxygen correlated significantly with S and N (p b 0.05 and p b 0.05 respectively) (Table 5).

4. Discussion The initial condition of benthic communities in the study area was degraded by the influence of the Athens Waste Water Treatment Plant (WWTP) sewage effluents before the dredged-material disposal. Dumping operations caused a reduction in the number of species and individuals but no significant changes in the ratio among tolerant and sensitive species. The dominance of tolerant species before dumping operations plays an important role in ecosystem response and recovery (Bolam, 2011). The results show that all stations were affected by dumping operations, but impacts and recovery patterns can be distinguished clearly into two groups of stations: Station 1 in the spoilground which is influenced by direct burial and stations outside the spoil-ground, affected by drift of the spoil and smothering. Station 1 has been severely affected by direct burial, and the communities were altered from a condition with the most abundant and

Fig. 3. ANOVA mean plots (95% LSD Intervals) of indices with significant change at the second group (Stations 2, 3, 4, 5). a) Number of Species (S), p b 0.05; b) Abundance (N), p b 0.05.

N. Katsiaras et al. / Science of the Total Environment 508 (2015) 320–330

327

Fig. 4. Cluster dendrogram and MDS plot of stations outside the spoil-ground (Stations 2, 3, 4, 5) per sampling periods. MDS plot was clustered in a similarity over 50% and it also shown with projected levels of sediment toxicity levels mg l−1. Pr: prior dumping, D: During dumping, Po: post to dumping.

diverse fauna to an almost azoic state during dumping. Apart from the frequent discharge, which did not permit the infauna to migrate vertically towards the seafloor; both strong grain-size dissimilarity and different chemical contents of the dredged sediments could have weakened the survival and re-colonization ability of species. Similar findings are also reported after deposition of simulated dredgedmaterial in the North Sea (Bolam, 2011). The functional group of species plays an important role in burial survival and re-colonization time and indeed the majority of the 11 species recorded during and post to dumping operations were motile carnivores. Carnivores are known to have sufficient ability of vertical migration (Bolam, 2011) and less dependence on sediment for shelter and food (Oug et al., 1998). The observed degradation of macrofauna communities caused by burying remains significant and greatly apparent post to dumping.

Although Station 2 is located within the dumping licensed area, the demonstrated impacts differ greatly from Station 1, and show smothering influence. It had significant spatial differences with the spoil-ground of Station 1 and it was clustered together with the outer stations. This is due to discharge of dredged-material only at Station 1, rather than spreading throughout the designated area. However, only post to dumping, there was a significant alteration of sediment grain-size towards fine-grained and a sudden increase of contaminant concentrations, combined with a degradation of number of species and abundance. These observations indicate a late discharging event at Station 2 at some point between the last sampling interval during dumping (October 2011) and the cease of the operations (January 2012). During dumping operations, there were also significant changes in the number of species and abundance in the outer stations. In these

328

N. Katsiaras et al. / Science of the Total Environment 508 (2015) 320–330

Table 5 Spearman Rank Correlations between biotic, sediment toxicity and environmental variables for stations outside the spoil-ground. Statistical significant correlations are indicated with bold. (r): correlation coefficient, (n): pairs of data values, (p): p-value.

Cu

Zn

AH

PAH

D.O.

PO4

Microtox SPT EC(50)

(r) (n) (p) (r) (n) (p) (r) (n) (p) (r) (n) (p) (r) (n) (p) (r) (n) (p) (r) (n) (p)

S

N

J

H

Bentix

Microtox SPT EC50

−0.1122 38 0.4948 −0.1765 38 0.2829 −0.3945 38 0.0164 −0.3011 38 0.067 0.3296 38 0.045 −0.1592 38 0.3329 0.4435 28 0.0212

−0.0312 38 0.8493 −0.1554 38 0.3445 −0.3866 38 0.0187 −0.2667 38 0.1047 0.3343 38 0.042 −0.2633 38 0.1092 0.3364 28 0.0804

0.0849 38 0.6057 0.2589 38 0.1152 0.1753 38 0.2864 0.1845 38 0.2619 −0.2231 38 0.1747 0.302 38 0.0662 0.0779 28 0.6857

−0.1367 38 0.4055 −0.0859 38 0.6014 −0.3742 38 0.0228 −0.2383 38 0.1471 0.1239 38 0.4511 −0.0367 38 0.8234 0.4168 28 0.0303

−0.2131 38 0.1949 −0.1913 38 0.2446 0.0294 38 0.8582 −0.1968 38 0.2313 −0.1757 38 0.2852 0.4351 38 0.0081 0.7153 28 0.0002

−0.4044 28 0.0356 0.6044 28 0.0017 −0.3934 28 0.0409 −0.5341 28 0.0055 0.0879 28 0.6477 0.3078 28 0.1098 – – –

stations there were no significant changes of sediment grain-size, therefore the observed degradation could not be linked to habitat alteration to a finer-grained substratum. The above diversity components, significantly correlated with bottom layer dissolved oxygen, concentrations of hydrocarbons and toxicity levels. The link of these factors with the dredged-material is confirmed also by the results of Panagiotopoulos et al. (2010), who investigated the sediment quality at the estuary of Kifissos River before operations. This indicates impacts from contaminants spread to the area, because of re-suspension, drift and smothering in an adjacent region at least 3.2 km away (Station 4) from the spoilground. The observed impacts had no significant spatial differences between the outer stations, during the 2 years of dumping operations. In general, there was no linear degradation with time; rather unstable conditions in macrofauna communities were observed throughout the dumping operations. This could be attributed to a non-stable direction of the dredged-material drift, since Saronikos Gulf seawater circulation pattern is complex and changes from cyclonic to anticyclonic depending on season (Kontoyiannis, 2010). Decreases of dissolved oxygen away from the spoil-ground could also be linked to the local circulation and delayed decomposition. A similar pattern was also observed by Pavlidou et al. (2012) in Saronikos Gulf, finding the lowest dissolved oxygen concentration as far as 14 km away from the pollution source. The outer stations, showed trends of recovery for species number and abundance post to dumping. In general, they were still more degraded in comparison to the condition prior to dumping, but this difference was not statistically significant. However, the aliphatic and polycyclic aromatic hydrocarbon concentrations and toxicity levels were found to be significantly high in the study area even post to dumping. In coastal marine communities, hydrocarbon concentrations over 100 μg g−1 are considered to be clear evidence of oil pollution (Hatzianestis et al., 2011) and sediment toxicity accepted limit for the Microtox® SPT test is set below 1000 mg l− 1 (Bombardier and Bermingham, 1999; Environment Canada, 2002). In the outer stations, AH concentrations were found up to 702.3 mg/g, PAH concentrations up to 1005.2 ng/g and for the Microtox® SPT test EC(50) values down to 528 mg/l post to dumping (Station 5 in all cases). Since sediment toxicity was found in combination with chemical contamination, it is an indication that the toxic contaminants are bio-available (Chapman, 1996) and the macrofauna communities of the study area are under stress even post to dumping. In this case, the observed recovery patterns in diversity components should not rule out the possible ecological hazard

that dumping operations caused in the area. The potential hazard tends to differ among species, since the ability to metabolize PAH varies (Driscoll and McElroy, 1996). The contaminant response can be reflected only in particular species or group of species, but this can precede more fundamental changes (Oug et al., 1998). More importantly, transfer to higher trophic levels should be considered, since even low PAH concentrations in sediments do not rule out bioaccumulation (Froehner et al., 2011). Re-suspension of contaminated sediments by dredging and dumping operations has been linked to community-level and sub-lethal responses in exposed populations of invertebrates and fish (Roberts, 2012). Knott et al. (2009) reported a severe decrease in the recruitment of sessile invertebrates within an estuary exposed to dredging and deposition of contaminated sediments (Port Kembla, Australia) and argued that these impacts were likely to be caused by the resuspension of the contaminated sediments into the water-column and particularly by the contaminants in, or released from, the sediments. Among ecological indicators, species number and abundance were considered by Ware et al. (2008) to be the most sensitive to dumping impacts, which is consistent with the present study. The number of species and abundance was proved also more sensitive in reflecting the initial impact and recovery process of the benthic communities after an accidental oil-spill in the Santorini caldera (Simboura et al., 2012) than the Shannon–Wiener diversity and the Bentix index which rather demonstrated the final stages of recovery. In some other study cases no relation was found between sediment contaminant concentrations and macrofauna community alterations; the degradations observed were attributed to physical disturbance and recovered (e.g. Stronkorst et al., 2003; Bolam et al., 2011). A long term monitoring of the effects of dumping coarse metalliferous waste on the benthic communities of the Northern Evoikos gulf (Greece), also showed that the impact was mainly confined to physical disturbance and instability of the sediment (Simboura et al., 2007), but no toxicity bioassays were performed. However, bioaccumulation levels were proved to significantly increase in benthopelagic organisms of the dumping area (Simboura and Catsiki, 2009). In that case, Bentix demonstrated more efficiently the long term degradation of the benthic communities, in comparison with the diversity and species richness indices. The response of a given metric varies to different stressors (Ware et al., 2008), since each metric has been developed using known species

N. Katsiaras et al. / Science of the Total Environment 508 (2015) 320–330

responses to specific impacts (i.e. sediment instability, chemical pollution, organic enrichment or general environmental stress). Our results confirm the need for investigating dumping impacts and recovery patterns of benthic macrofauna with the use of sediment chemical analysis and toxicity bioassays, as it is stressed in the past by other authors (e.g. Chapman, 1996; Chapman and Hollert, 2006; references within). In the present study, where the dredged-material was highly contaminated by aliphatic and polycyclic aromatic hydrocarbons, the macrofauna responses were found to correlate with sediment toxicity. In addition, taking only ecological indicators into account may not be enough to determine the full extent of impacts on the ecosystem caused by dumping in this case. 5. Conclusions • Dumping operations imposed significant direct (burial) and indirect (smothering) effects to the benthic communities. • Direct impacts caused a severe alteration from a diverse and abundant community to an almost azoic state. Post to dumping, the degradation around the spoil-ground remained greatly apparent. • Indirect impacts to the benthic communities were reflected mostly in species number, abundance and toxicity levels. Four months after ceasing of operations, some recovery patterns were shown in terms of species number and abundance, but toxicity levels remained elevated. • High concentrations of hydrocarbons and levels of toxicity found post to dumping show that there are bio-available contaminants in the area and the benthic communities remain under stress. • Apart from macrofauna communities, sediment chemical analyses and toxicity bioessays are needed to determine the full extent of ecological impacts caused by dumping.

Acknowledgments This study was done under Sarge Project, which was funded by the Ministry of the Environment, Physical Planning and Public Works EYDE/AELMP/241/FE14/22.04.2010). We would like to thank our colleague Anastasia Lambou for designing the study area map. References Azur Environmental. Microtox® Test Manual, Carlsbad, USA; 1998. Bianchi GG, Hall IR, McCave IN, Joseph L. Measurement of the sortable silt current speed proxy using the Sedigraph 5100 and Coulter Multisizer IIe: precision and accuracy. Sedimentology 1999;46:1001–14. Bolam SG. Burial survival of benthic macrofauna following deposition of simulated dredged material. Environ. Monit. Assess. 2011;181:13–27. Bolam SG, Rees HL. Minimizing impacts of maintenance dredged material disposal in the coastal environment: a habitat approach. Environ. Manag. 2003;32(2):171–88. Bolam SG, Barry T, Bolam T, Mason C, Rumney HS, Thain JE, Law JR. Impacts of maintenance dredged material disposal on macrobenthic structure and secondary productivity. Mar. Pollut. Bull. 2011;62:2230–45. Bombardier M, Bermingham N. The SED-TOX index: toxicity-directed management tool to assess and rank sediments based on their hazard—concept and application. Environ. Toxicol. Chem. 1999;18:685–98. Botsou F, Hatzianestis I. Polycyclic aromatic hydrocarbons (PAHs) in marine sediments of the Hellenic coastal zone, Eastern Mediterranean: levels, sources and toxicological significance. J. Soils Sediments 2012;12:265–77. Brannon JM, Poindexter-Rollings ME. Consolidation and contaminant migration in a capped dredged-material deposit. Sci. Total Environ. 1990;91:115–26. Carpenter JH. The accuracy of the Winkler method for the dissolved oxygen analysis. Limnol. Oceanogr. 1965;10:135–40. Chapman PM. The sediment quality triad approach to determining pollution-induced degradation. Sci. Total Environ. 1990;97(98):815–25. Chapman PM. Presentation and interpretation of sediment quality triad data. Ecotoxicology 1996;5:327–39. Chapman PM, Hollert H. Should the sediment quality triad become a tetrad, pentad, or possibly even a hexad? J. Soils Sediments 2006;6(1):4–8. Driscoll SK, McElroy AE. Bioaccumulation and metabolism of benzo(a)pyrene in three species of polychaete worms. Environ. Toxicol. Chem. 1996;15:1401–10.

329

Environment Canada. Biological test method: reference method for determining the toxicity of sediment using luminescent bacteria in a solid-phase test. Report EPS 1/RM/42. Ottawa: Environmental Protection Service; 2002. Essink K. Ecological effects of dumping of dredged sediments: options for management. J. Coast. Conserv. 1999;5:69–80. European Commission. Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000, establishing a framework for community action in the field of water policy. Off. J. Eur. Union 2000;L327:1–72. Fauchald K, Jumars PA. The diet of worms: a study of polychaete feeding guilds. Oceanogr. Mar. Biol. Annu. Rev. 1979;17:193–284. Folk RL. Petrology of Sedimentary Rocks. Austin, Texas: Hemphill Publ. Co.; 1974. Froehner S, Maceno M, Machado KS. Predicting bioaccumulation of PAHs in the trophic chain in the estuary region of Paranagua. Environ. Monit. Assess. 2011;174:135–45. GIG (Geographical Intercalibration Group). Technical report coastal waters, Mediterranean GIG Benthic Invertebrate fauna, JRC and the benthic macroinvertebrates subgroup. https://circabc.europa.eu/faces/, 2013. Grall J, Glémarec M. Using biotic indices to estimate macrobenthic community perturbations in the bay of Brest. Estuar. Coast. Shelf Sci. 1997;44:43–53. Harvey M, Gauthier D, Munro J. Temporal changes in the composition and abundance of the macro-benthic invertebrate communities at dredged material disposal sites in the Anse h Beaufils, Baie des Chaleurs, Eastern Canada. Mar. Pollut. Bull. 1998;36(1): 41–55. Hatzianestis I, Plakidi E, Pyrgaki X. Hydrocarbons. In: Kapsimalis V, editor. Environmental monitoring of the marine dumping area (Saronikos Gulf) of the dredged material from the low course of Kifissos river. Technical ReportGreek: Hellenic Centre for Marine Research; 2011. p. 141–57. Kapsimalis V, Panagiotopoulos I, Kanellopoulos Th, Hatzianestis I, Antoniou P, Anagnostou Ch. A multi-criteria approach for the dumping of dredged material in the Thermaikos Gulf, Northern Greece. J. Environ. Manag. 2010;91:2455–65. Karageorgis AP, Anagnostou CL, Kaberi H. Geochemistry and mineralogy of the NW Aegean Sea surface sediments: implications for river runoff and anthropogenic impact. Appl. Geochem. 2005;20:69–88. Knott NA, Aulbury JP, Brown TH, Johnston EL. Contemporary ecological threats from historical pollution sources: impacts of large-scale resuspension of contaminated sediments on sessile invertebrate recruitment. J. Appl. Ecol. 2009;46:770–81. Kontoyiannis H. Observations on the circulation of Saronikos Gulf: a Mediterranean embayment sea-border of Athens, Greece. J. Geophys. Res. 2010;115. (23 pp.). Morton JW. Ecological effects of dredging and dredge spoil disposal: a literature review. Technical papers of the US Fish and Wildlife Service, 94. Technical paper; 1977. (33 pp.). OSPAR Commission. Literature review on the impacts of dredged sediment disposal at sea. Biol. Divers. Ecosyst. 2008;362:54pp. Oug E, Naes N, Rygg B. Relationships between soft bottom macrofauna and polycyclic aromatic hydrocarbons (PAH) from shelter discharge in Norwegian fjords and coastal waters. Mar. Ecol. Prog. Ser. 1998;173:39–52. Panagiotopoulos I, Kapsimalis V, Hatzianestis I, Kanellopoulos T, Kyriakidou C. Environmental status of the metropolitan river (Kifissos) of Athens, Greece. Environ. Earth Sci. 2010;61:983–93. Pavlidou A, Hatzianestis I, Psillidou-Giouranovits R. Dissolved oxygen decrease near the bottom of Inner Saronikos Gulf affected by the Athens Sewage Outfall. EGU General Assembly 2012, 22–27 April, 2012 in Vienna, Austria. Vol. 14; 2012. p. 4652. Roberts DA. Causes and ecological effects of resuspended contaminated sediments (RCS) in marine environments. Environ. Int. 2012;40:230–43. Roberts RD, Forrest BM. Minimal impact from long-term dredge spoil disposal at a dispersive site in Tasman Bay, New Zealand. J. Mar. Freshw. Res. 1999;33:623–33. Simboura N, Catsiki VA. The environmental impact of discharging metalliferous waste in a Mediterranean coastal area (Aegean Sea, N. Evvoikos gulf, Greece). In: McLaughlin NY, Breaux LA, editors. Chemical Mineralogy, Smelting and Metallization. Nova Science Publishers Inc.; 2009. p. 215–23. Simboura N, Nicolaidou A. The Polychaetes (Αnnelida, Polychaeta) of Greece: checklist, distribution and ecological characteristics. Monographs on Marine Sciences, Series No 4NCMR; 2001. Simboura N, Zenetos A. Benthic indicators to use in ecological quality classification of Mediterranean soft bottom marine ecosystems, including a new biotic index. Mediterr. Mar. Sci. 2002;3(2):77–111. Simboura N, Panayotidis P, Papathanassiou E. A synthesis of the biological quality elements for the implementation of the European Water Framework Directive in the Mediterranean ecoregion: the case of Saronikos Gulf. Ecol. Indic. 2005;5(3):253–66. Simboura N, Papathanassiou E, Sakellariou D. The use of a biotic index (Bentix) in assessing long term effects of dumping coarse metalliferous waste on soft bottom benthic communities. Ecol. Indic. 2007;7(1):164–80. Simboura N, Pancucci-Papadopoulou MA, Reizopoulou S, Streftaris N, Arvanitakis G, Hatzianestis J. The impact of the cruise ship Sea Diamond wreckage on the Santorini island (Aegean Sea, Eastern Mediterranean) Caldera benthic ecosystem. 10th Hellenic Symposium of Oceanography & Fisheries. 7–11 May 2012, Athens; 2012. p. 228. (Book of Abstracts). Simboura N, Zenetos A, Pancucci-Papadopoulou MA. Benthic community indicators over a long period of monitoring (2000–2012) of the Saronikos Gulf, Greece, Eastern Mediterranean. Environ. Monit. Assess. 2014;186:3809–21. Simonini R, Ansaloni I, Cavallini F, Graziosi F, Iotti M, Massamba N'Siala G, Mauri M, Montanari G, Preti M, Prevedelli D. Effects of long-term dumping of harbourdredged material on macrozoobenthos at four disposal sites along Emilia-Romagna coast (Northern Adriatic Sea, Italy). Mar. Pollut. Bull. 2005;50:768–77. Smith SDA, Rule MJ. The effects of dredge-spoil dumping on a shallow water softsediment community in the Solitary Islands Marine Park, NSW, Australia. Mar. Pollut. Bull. 2001;42(11):1040–8.

330

N. Katsiaras et al. / Science of the Total Environment 508 (2015) 320–330

Stronkorst J, Ariese F, Van Huttum B, Postma JF, De Kluijver M, Den Besten PJ, Bergman MJN, Daan R, Murk AJ, Vethaak AD. Enviromental impact and recovery at two dumping sites for dredged material in the North Sea. Environ. Pollut. 2003;124:17–31. Toumasiz AD. Enviromental impact associated with the dumping of dredged material at sea. A study for the Limassol port extension works. Water Sci. Technol. 1995; 32(9–10):151–8. Underwood AJ. Beyond BACI: experimental designs for detecting human environmental impacts on temporal variations in natural populations. Aust. J. Mar. Freshw. Res. 1991;42:569–87. UNEP. Guidelines for the management of the dredged-material. UNEP(OCA)/MED IG.12/ 4; 1999.

Ware SJ, Rees HL, Boyd SE, Birchenhough SN. Performance of selected indicators in evaluating the consequences of dredged material relocation and marine aggregate extraction. Ecol. Indic. 2008;9(4):704–18. Wilber DH, Clarke D, Rees SI. Responses of benthic macroinvertebrates to thin-layer disposal of dredged material in Mississippi Sound, USA. Mar. Pollut. Bull. 2007;54:42–52. Witt J, Schroeder A, Knust R, Arntz WE. The impact of harbour sludge disposal on benthic macrofauna communities in the Weser estuary. Helgol. Mar. Res. 2004;58:117–28. Zimmerman LE, Jutte PC, Van Dolah RF. An environmental assessment of the Charleston Ocean dredged material disposal site and surrounding areas after partial completion of the Charleston harbor deepening project. Mar. Pollut. Bull. 2003;46:1408–19.