Integrated ecotoxicological assessment of marine ... - Springer Link

Ecotoxicology (2013) 22:996–1011 DOI 10.1007/s10646-013-1085-6

Integrated ecotoxicological assessment of marine sediments affected by land-based marine fish farm effluents: physicochemical, acute toxicity and benthic community analyses C. Silva • E. Yań˜ez • M. L. Martıń-Dıáz I. Riba • T. A. DelValls

•

Accepted: 2 May 2013 / Published online: 17 May 2013 Ó Springer Science+Business Media New York 2013

Abstract An integrated ecotoxicological assessment of marine sediments affected by land-based marine fish farm effluents was developed using physicochemical and benthic community structure analyses and standardised laboratory bioassays with bacteria (Vibrio fischeri), amphipods (Ampelisca brevicornis) and sea urchin larvae (Paracentrotus lividus). Intertidal sediment samples were collected at five sites of the Rio San Pedro (RSP) creek, from the aquaculture effluent to a clean site. The effective concentration (EC50) from bacterial bioluminescence and A. brevicornis survival on whole sediments and P. lividus larval developmental success on sediment elutriates were assessed. Numbers of species, abundance and Shannon diversity were the biodiversity indicators measured in benthic fauna of sediment samples. In parallel, redox potential, pH, organic matter and metal levels (Cd, Cu, Ni, Pb and Zn) in the sediment and dissolved oxygen in the interstitial water were measured in situ. Water and sediment physicochemical analysis revealed the exhibition of a spatial gradient in the RSP, evidenced by hypoxia/anoxia, reduced and acidic conditions, high organic enrichment and

C. Silva M. L. Martıń-Dıáz I. Riba T. A. DelValls UNITWIN/UNESCO/WiCoP, Physical Chemical Department, Campus de Excelencia Internacional del Mar (CEIMAR), University of Ca´diz, Polı´gono Rıó San Pedro s/n, 11510 Ca´diz, Puerto Real, Spain C. Silva M. L. Martıń-Dıáz Andalusian Center of Marine Science and Technology (CACYTMAR), CEIMAR, University of Ca´diz, Polı´gono Rıó San Pedro s/n, 11510 Ca´diz, Puerto Real, Spain C. Silva (&) E. Yań˜ez School of Marine Science, Pontificia Universidad Cato´lica de Valparaı´so, Avda. Altamirano, 1480 Valparaı´so, Chile e-mail: [email protected]

123

metal concentrations at the most contaminated sites. Whereas, the benthic fauna biodiversity decreased the bioassays depicted decreases in EC50, A. brevicornis survival, P. lividus larval success at sampling sites closer to the studied fish farms. This study demonstrates that the sediments polluted by fish farm effluents may lead to alterations of the biodiversity of the exposed organisms. Keywords Sediment organic enrichment Microtox Amphipod survival Sea urchin larval development Benthic fauna Aquaculture effluents

Introduction Marine fish aquaculture has become more intensive in recent years, mainly as a result of the introduction of new technologies, expansion of suitable sites, improvements in feed technology, better understanding of the biology of the species farmed, improved water quality within farming systems and increased demand for fish products (Read and Fernandes 2003). However, this strong expansion in the marine fish aquaculture industry has brought significant marine environmental impacts to coastal ecosystems, including sediment organic enrichment and eutrophication (Holmer et al. 2005; Kalantzi and Karakassis 2006); chemical pollution from pharmaceuticals, organics, bactericides and metals (Antunes and Gil 2004; Cabello 2006; Sapkota et al. 2008); and changes in the biodiversity and community structures of benthic fauna (Vezzulli et al. 2008; Tomassetti et al. 2009). As an example, Rio San Pedro (RSP) is a salt marsh creek situated in south-western Spain and has traditionally been a highly productive area for aquaculture. Untreated aquaculture wastewaters from a land-based marine fish

Integrated ecotoxicological assessment of marine sediments

farm (LBMFF) in the upper part of the creek are discharged directly into the coastal waters, constituting a major source of pollution in the surroundings. Organic enrichment contamination is produced by waste in dissolved and particulate forms, and although most of the carbon that is fed to the fish is converted into biomass, a considerable amount of unconsumed food and faeces settle out as sediment (Karakassis et al. 2002; Papageorgiou et al. 2009). The area affected by the organic enrichment of the fish farm effluent in the RSP is characterised by low pH and high levels of nutrients, particulate organic matter (POM), suspended solids and metals (Tovar et al. 2000a; Mendiguchıá et al. 2006; De la Paz et al. 2008b). Organic enrichment may change the physical and chemical compositions of sediments (Karakassis et al. 2002), affecting the structure of the macrobenthic communities and the health status of the biota (Solan et al. 2004). Most studies of effluents from LBMFFs have been focused on chemical contamination and benthic fauna impacts (Tello et al. 2010). However, to assess the environmental impacts of fish aquaculture discharges, laboratory bioassays should be incorporated in integrated analysis, as these toxicity tests have been proven as useful and cost-effective tools to evaluate marine water and sediment samples from contaminated sites (Morales-Caselles et al. 2007; Carballeira et al. 2012). An integrated approach for different biomonitoring levels using analysis of chemical contamination, determination of benthic community structure and assessment of toxicity using laboratory bioassays was proven to be useful for establishing evidence of exposure to pollutants and damage to the health of sentinel organisms (Cesar et al. 2009). Standardised toxicity tests are useful in coastal ecosystem management and offer several advantages over chemical analyses because they provide information about the bioavailability and toxicity of metals and other contaminants to organisms (Peters et al. 2002; Hernando et al. 2007). Marine bioassays have occasionally been used to evaluate the impact of cagebased fish aquaculture (Marıń et al. 2007) and LBMFFs on water contamination (Hernando et al. 2007; Carballeira et al. 2011a, b, 2012). Ecological monitoring using benthic structure analysis will provide information on changes in species composition, abundance and diversity, all of which may be indicative of the effects of pollution in communities (Chapman, 1996; DelValls et al. 1998a). However, scarce information is available concerning the use of an integrated approach of physicochemical, bioassays and benthic alteration analyses to assess the potential effects of sediment affected by these activities. Different toxicity tests using different test matrices (water and sediment) and trophic levels can be applied for the assessment of marine sediments. Bioassays studies on sediment use whole sediment or aqueous extracts (e.g.

997

elutriate). Whole sediment bioassays (e.g. bacteria and amphipods) provide information on particle associated contaminants, while the elutriate toxicity test (e.g. sea urchin larval development) provides information on the water-soluble contaminants. The Microtox bioassay is a bioluminescence inhibition test that uses the bacteria Vibrio fischeri and has been widely applied to evaluate the toxicity of contaminated sediments (Casado-Martinez et al. 2006a; Morales-Caselles et al. 2007). Previous studies have shown that the amphipods species successfully used in routine toxicity tests of marine sediment include Corophium volutator (Morales-Caselles et al. 2007), Corophium multisetosum (Montero et al. 2011), Ampelisca brevicornis (Riba et al. 2003; Ramos-Go´mez et al. 2009) and Microdeutopus gryllotalpa (DelValls et al. 1999). Additionally, the sea urchin embryo development test is a cost-effective and useful standard bioassay to assess the toxicity of contaminated water and sediments (Beiras et al. 2003; Cesar et al. 2009; Carballeira et al. 2012). This study presents an integrated biomonitoring approach for studying the ecotoxicity of marine sediments affected by LBMFF effluents using physicochemical and benthic community structure analyses and standardised laboratory bioassays with bacteria (V. fischeri), amphipods (A. brevicornis) and sea urchin larvae (Paracentrotus lividus). The aims were to evaluate the suitability of a battery of acute toxicity tests at different trophic levels for sediment quality assessments and to study the associations between physicochemical characteristics, bioassays responses and benthic fauna indicators. This approach will guarantee an improved biomonitoring for environmental quality and ecological risk assessments in coastal marine ecosystems (DelValls et al. 1999; Chapman et al. 2002).

Methodology Study area The study was carried out in the RSP (36°230 –370 N, 6°80 – 150 W) (Fig. 1). The RSP is a shallow tidal creek within the salt marsh area of the Bay of Cadiz (SW Spain). The area was selected because a LBMFF facility is located in the upper part of the creek. The RSP creek used to be a tributary of the Guadalete River until it was artificially blocked at 12 km from the mouth; currently, the only freshwater input into the creek comes from precipitation and land drainage. The landscape surrounding the RSP was originally formed by an extensive area of salt marshes, progressive exploitation by the human population, such as salt marsh desiccation by blockage, fish farm construction, salt production facilities, and other anthropogenic activities, has significantly reduced the proportion of the area as

123

998

C. Silva et al.

Fig. 1 Location of study sites in the Rıó San Pedro creek. The map shows the main land uses. Points with labels in the marine coastal area correspond to sampling locations and the control site (CA) for the amphipods bioassay

natural marsh (Ferroń et al. 2009). The actual RSP creek is effectively isolated, laterally by means of an embankment that separates the creek from the various industries that exploit the salt marsh. This human-made separation suggests that the influence of the salt marsh on the RSP is only moderate (De la Paz et al. 2008b; Ferroń et al. 2009). The RSP creek holds complementary and in some cases conflicting uses, including recreational shipping, fishing boats anchorages, semi-intensive marine fish aquaculture and protected natural areas (Fig. 1). A LBMFF is located in the upper part of the creek, producing 550 tons year-1 of gilthead sea bream (Sparus aurata) and European sea bass (Dicentrarchus labrax) (Empresa Pu´blica Desarrollo Agrario y Pesquero, Consejerıá de Agricultura y Pesca). The fish are fed with manufactured feeds with a nutrient composition of 45 % of protein, 14 % of lipids and 20 % of carbohydrates. Point source discharges of effluents from the fish farm (Fig. 1) are discharged into the creek and constitute a major source of pollution and deterioration of the environmental quality in the RSP creek. The total volume of water introduced daily by the fish farm ranges from 180,000 to 290,000 m3 (Tovar et al. 2000b). The LBMFF consists of several batteries of shallow earthen ponds excavated in the soil. The overlying water, enriched

123

with the metabolic products coming from the organically enriched contamination produced by the waste of the fish farm, is directly discharged to the upper part of the creek. Tovar et al. (2000a) estimated the total amount of dissolved nutrients, total suspended solids (TSS), POM and biochemical oxygen demand (BOD) that the fish farm discharged into the receiving waters for each ton of cultured fish. According to this, approximately 9.1 9 106 kg TSS, 8.4 9 105 kg POM, 2.4 9 105 kg BOD, 36 9 103 kg N– NH4?, 5.0 9 103 kg N–NO2-, 6.7 9 103 kg N–NO3- and 2.6 9 103 kg P–PO43- are discharged annually into the environment. As a consequence, the concentration of nutrients and suspended solids in the creek is relatively high. The RSP is subject to a semi-diurnal tidal regime with the height of the tide varying from 3.5 m at spring tide to 0.5 m at neap tide. The creek has a maximum depth of 4–5 m and water input comes from tides from the Bay of Cadiz making water renewal in the upper part (aquaculture zone) to be very poor. Tovar et al. (2000a) studied the longitudinal distribution of various physicochemical properties in the RSP and suggested that the low pH values in the fish farm effluent are due to the high ammonium concentration and to the acidic character of the faeces and the industrial fish feed. The seasonal pattern of nutrients and


other physicochemical parameter in the RSP is different from the known seasonal pattern for coastal waters (Tovar et al. 2000a), and it follows the growth rate curve of the fish cultivated on the farm (Tovar et al. 2000b). The chlorophyll a value follows this trend in response to high nutrient inputs, mainly nitrites and nitrates, and to the favourable conditions caused by residence time (De la Paz et al. 2008b). The fish farm is large enough to be considered the main source of dissolved nutrients, POM, and TSS to the system. Tovar et al. (2000a) and De la Paz et al. (2008b) identified the existence of two different zones of water quality within the inlet: the first, with a length of approximately 8 km, is closer to the mouth, has no fish farms and has good water renovation, controlled by the tides; the second section is affected by the semi-intensive fish culture, and its effluents are considered the main source of nutrients, organic matter, and suspended solids for the ecosystem. No bioassay studies were made in the RSP creek. De la Paz et al. (2008b) observed a strong seasonality of changes in dissolved inorganic carbon, pH, and dissolved oxygen concentrations. These changes were associated with an increase in metabolic rates with temperatures, the alternation of storm events and high evaporation rates and an intense seasonal variability in the discharges from fish farms. De la Paz et al. (2008a) suggested that the seasonal variability of water-saturated CO2 fugacity is related to the seasonal variability in discharges from the fish farm and to the increase with temperature of organic matter respiratory processes in the tidal creek, and values observed are in the same range as several highly polluted European estuaries or waters surrounding mangrove forests. These authors also concluded that the RSP acts as a source of CO2 to the atmosphere throughout the year, with the summer accounting for the higher average monthly flux. Ferroń et al. (2009) estimate that approximately 60 % of the total POM that is discharged into the creek by the main fish farm facility is estimated to degrade in the sediments, resulting in a significant input of nutrients to the system. Sampling strategy Samplings were carried out in the spring (May) of 2011 during low tide. A total of five sampling sites (RSP1, RSP2, RSP3, RSP4 and RSP5) were established at 0.05, 0.3, 2.5, 4.3 and 9.5 km, respectively, following a presumably gradient of contamination from the aquaculture effluent to the reference site (RSP5) (Fig. 1). The latter is located in the lower part of the creek, in an area that is not affected by the fish farm. RSP5 was considered as a reference site due to its good water renovation (Tovar et al. 2000a), and as it has been previously used as a control zone for biomonitoring studies (Pe´rez et al. 2004; Sole´ et al. 2009).

999

At each site, water samples were collected in five replicates at one meter depth for in situ measurement of temperature (T) and salinity (SA). Sediment samples for chemical analysis and toxicity tests were collected intertidally at each sampling site using a 0.025 m2 Van Veen grab and three samples were taken at each site. After collection, the sediments were kept in a cooler at 4 °C and transported to the laboratory. Sediments were sieved through a 2-mm mesh and stored in a cooler (4 °C) for 2 weeks in darkness until the toxicity tests were carried out. Before use, sediment was homogenized with a Teflon spoon until no color or textural differences were detected. All beakers were thoroughly cleaned with acid (10 % HNO3) and rinsed in double de-ionized (Milli-Q) water prior to sampling and storage. The samples were divided for chemical analysis and toxicity tests. To minimize the effect of grain size on metal distribution, the analyses were carried out in the fine (\63 lm) portion of sediment (Usero et al. 2008). The granulometric fraction was separated from the samples with the manual wet sieving method using Nylon sieves and Milli-Q water. Sediment samples for benthic fauna analysis were taken with the Van Veen grab in the lower intertidal sediment at low tide. Three samples were taken at each site to determine the variability between samples. All samples were sieved using a 0.5 mm mesh size sieve and retained fauna were fixed in 10 % buffered formalin. After 3 days the samples were preserved in 70 % ethanol until the benthic community analysis was performed. Physicochemical analysis The water T was registered using a VWR CO3000H electrode (ref. 663–0,143) with an accuracy of ±0.1 °C. The water SA was measured with a VWR CO3000H electrode with an accuracy of ±0.01 psu (practical salinity units), which was previously calibrated with a control standard solution of 0.01 mol L-1 of KCl. Redox potential (Eh) and pH were measured immediately after collection in sediment samples at a depth of 3 cm. The Eh was measured (accuracy of ±0.1 mV) with a VWR Pt electrode (ref. 662–9,906) that was previously calibrated with a Hamilton Redox Buffer standard (Cat. No 1,412,060) of known mV value (475 mV at 25 °C). The pH was measured with a VWR ph1000H electrode (ref. 662–1,151) that was previously calibrated with two buffers (pH = 7 and 10) with an accuracy of ±0.001 units. Dissolved oxygen (DO) in the interstitial water of the sediments was determined using a VWR OX4000H electrode (ref. 664–0,038) with an accuracy of 0.001 mg L-1. Sediment samples were stored and dried at room temperature until they were processed for measuring the total organic matter (TOM) content, which was obtained through calcination (loss of ignition) in a

123

1000

muffle furnace at 450 °C for 5 h with an accuracy of ±0.1 %. Metals were determined in fine (particles \63 lm) fraction of dried sediment samples: cadmium (Cd), copper (Cu), nickel (Ni), lead (Pb) and zinc (Zn). Metal concentrations were determined using the method of aqua regia extraction (ISO11466 1995). The overnight pre-digestion consisted on the addition of a mixture of aqua regia (10 mL Milli-Q water, 5 mL HNO3, 1 mL HCl) and 4 mL HF to a dried sediment aliquot (approx. 0.1 g) in a Teflon vessel. The samples were further digested in a Millestone Ethos 1,200 microwave oven to complete digestion. After cooling, the product of digestion were made up to 50 mL using Milli-Q water and kept in acid-treated plastic bottles. Analysis of metal concentrations (Cd, Cu, Ni, Pb and Zn) were carried out by inductively coupled plasma mass spectrometry (ICP-MS) using a Thermo Elemental X-7 equipment. All trace metal determinations were verified with the standard reference material MESS-1 NRC. Recovery of metals in the reference material was between 95 and 99 % of the certified values in the aqua regia extraction. The results were expressed as total concentration per lg g-1 dry weight. Standard quality assurance/ quality control (QA/QC) procedures included the use of method blanks, laboratory control samples, reference standards and calibration verifications. Bioassays Sediment potential toxicity under laboratory conditions were tested by three bioassays: the bacteria (V. fischeri) bioluminescence (i.e. Microtox), the amphipod (A. brevicornis) survival in the whole sediment and the embryological success of sea urchin (P. lividus) larvae in sediment elutriates. Bioluminescence test with bacteria The Microtox test with V. fischeri (NRRL-B-11177) measures the reduction in luminescence emitted by the bacteria when exposed to a contaminated matrix. The bioassay was conducted using Model 500 Analyser (SDI, USA) and was determined following the ISO 11348-1:2001 norm and the protocols of Environment Canada (2002) for the Basic Solid Phase Test (BSPT) (Azur Environmental 1998) adapted by Casado-Martıńez et al. (2006a). Briefly, 7 g (±0.01 g) of sediment were tested as suspensions prepared with 35 mL of commercial Microtox BSPT Diluent and diluted to a series of nine concentrations (dilutions) in the cuvettes. The reconstituted bacteria were added to the dilutions which were incubated for a period of 20 min at 15 °C in a waterbath. Next, the dilutions were filtered and

123

C. Silva et al.

500 mL of each cuvette of the bath were transferred to its corresponding cuvette in the apparatus and bioluminescence was measured in the ‘‘read well’’. Three replicates per sediment sampling site were considered. The reference site RSP5 was used as the control. The BSPT EC50 results were expressed as the effective concentration (mg sediment L-1) of whole sediments corrected for moisture content, which causes a 50 % reduction in bioluminescence following 5, 15 and 30 min of exposure. The criteria proposed in Spain to consider a sediment as toxic (EC50 \ 750 mg L-1 d.w. = 0.075 %) (Morales-Caselles et al. 2007) and the statistical difference with the control were used to classify the samples. Amphipod test Amphipods (A. brevicornis) were collected at low tide from a clean site (CA) in the Bay of Cadiz (Fig. 1) which was used as a control site for this bioassay. Organisms were separated and acclimatised in the laboratory (12 days) before the start of the bioassay. Following the Environment Canada (1992a) protocol adapted by Casado-Martıńez et al. (2006b), A. brevicornis were acclimatised in 11-L aquariums filled with sediment from the reference site and filtered and UV-sterilised natural seawater (salinity 32 psu) and they were maintained under controlled conditions (temperature, 20 ± 1 °C, pH, 7.8 ± 0.1, continuous aeration; 12 h-light:dark photoperiod). Because sexual classification requires extra manipulation that could lead to stress and body damage, the samplings were randomly made. It was assumed that the A. brevicornis used were representative of the natural population and organisms with a similar body size were selected. The bioassay was performed by exposing individual A. brevicornis to sediments from the different sampling sites. Three replicates per sediment sampling site (including the control site CA) were considered in the bioassay. The sediments were sieved (1 mm) prior to the toxicity test. Aliquots of 200 g of sediments were placed in 2-L glass beakers, and 800 mL of clean seawater was added. After the sediments settled down in the beakers, aeration was provided, and 12 h later, the organisms were sieved from the acclimatisation aquariums, and 20 A. brevicornis adults (3–5 mm) were placed in each replicate for 10 days of exposure. During the exposure period, constant aeration was supplied, and water quality parameters (temperature, salinity, dissolved oxygen and pH) and the photoperiod were controlled to maintain conditions similar to those of the acclimation phase. At the end of the test, the sediments were sieved (1 mm), and the A. brevicornis survival percentage was calculated in each replicate. The criterion of mortality was established as the total absence of movement and no response after repeated


mechanical stimulus. Samples were classified as toxic when survival was 20 % lower and significantly different from the control site CA (US EPA 1994; Riba et al. 2004; Casado-Martıńez et al. 2006b; Ramos-Go´mez et al. 2009). Sea urchin larval development test Adult sea urchins (P. lividus) were collected by scuba diving in an intertidal zone at a clean site on the coast of Andalucıá. Following collection, the P. lividus were immediately placed in a 30-L plastic cooler and kept at 18 °C with seawater from the same site for their transport to the laboratory. The P. lividus larval development bioassay followed the Environment Canada (1992b) protocol that was adapted by Beiras (2002) and Casado-Martıńez et al. (2006c). In the laboratory, organisms were maintained in aerated pools at 18 °C for acclimation during 10 days. Gametes were obtained by injecting 1 mL of KCl (0.5 M) through the peri-oral membrane of three specimens of each sex. Pools of eggs and sperm were prepared. Elutriates of sediment were obtained after 30 min of rotation (60 rpm) of the mixture of 100 g sediments and 500 mL of filtered and UV-sterilised natural seawater (salinity 32 psu). The mixtures were left to settle for 12 h at 20 °C in darkness. Afterwards, the overlying water was transferred to 20-mL incubation vials, and approximately 400 in vitro fertilised P. lividus eggs were placed in the vials. Three replicates were prepared for each elutriates sample and negative toxicity control samples consisting of clean seawater (CSW). The samples were incubated for 48 h at 20 °C and then the larvae development was stopped, and

1001

the larvae were fixed by adding a drop of formalin (40 %). The end point measured was the percentage of embryological success. One hundred larvae were counted using an optical microscope (OLYMPUS CKX41), considering normal larvae as well-developed four-arm embryos according to the skeletal abnormality criteria reported by Carballeira et al. (2011b). Samples were classified as toxic when the success of normally developed larvae was 20 % lower and significantly different from the control (CasadoMartıńez et al. 2006c). Integrated toxicity index In order to synthesise the three toxicity results into a single measure, a scoring system of 1–6 (Table 1) was developed based on the decision criteria for integrating individual toxicity data proposed by McPherson et al. (2008) and adapted by Alvarez-Guerra et al. (2009). The results of the A. brevicornis and P. lividus bioassays were assessed, using the percentage reduction in survival and success of normally developed larvae, respectively (i.e., whether the reduction was[20 % or[50 %, respectively), giving more weight to toxicity data that were statistically significant (P \ 0.05) compared to the reference site than to data that showed no statistically significant difference (McPherson et al. 2008). Regarding Microtox BSPT, the criteria proposed in Spain to consider a sediment sample as toxic (EC50 \ 750 mg L-1 d.w. = 0.075 %) were used within the scoring system as a reference threshold for assessing the Microtox results. The final toxicity index (TI) results were mapped.

Table 1 Score system used for integrating bioassays results into one toxicity index (based on McPherson et al. (2008) and adapted by AlvarezGuerra et al. (2009)) Toxicity index

Toxicity categories

Observed pattern in toxicity data

6

High

Greater than a 50 % reduction in at least one acute endpoint (i.e. survival and/or success)

5

High-moderate

4

Moderate

3

Low-moderate

2

Low

1

Non-toxic

Greater than a 20 % reduction in two acute endpoints and the differences are statistically significant Greater than a 20 % reduction in at least one acute endpoint (i.e. survival and/or success) with a statistically significant difference, and Microtox EC50 is \750 mg sediment L-1 Greater than a 20 % reduction in at least one acute endpoint (i.e. survival and/or success) with a statistically significant difference, and Microtox EC50 is [750 mg sediment L-1 Greater than a 20 % differences are not Greater than a 20 % differences are not

reduction both in amphipod survival and sea urchin success, but the statistically significant; Microtox EC50 is \750 mg sediment L-1 reduction both in amphipod survival and sea urchin success, but the statistically significant; Microtox EC50 is [750 mg sediment L-1

Greater than a 20 % reduction only in one acute endpoint (i.e. survival or success) but the differences are not statistically significant Less than a 20 % reduction both in amphipod survival and sea urchin success but Microtox EC50 is \750 mg sediment L-1 Less than a 20 % reduction both in amphipod survival and sea urchin success, and Microtox EC50 is [750 mg sediment L-1

123

1002

Benthic community analysis Samples were sorted and organisms were counted and identified to the lowest taxonomic level under a stereo microscope, using available species keys (Riedl 1986; Ocanã-Martıń et al. 2000). The biological material collected included both macrofauna and meiofauna (nematodes). The total species (S) value was calculated as the total number of species collected at each sampling site. The mean abundance (number of individuals per 0.025 m-2; N) and Shannon diversity (H’; Shannon and Weaver 1949) were also calculated for each sampling site. Benthic alteration index A benthic fauna communities alteration index was developed using the M-AMBI (Multivariate AZTI’s Marine Biotic Index) method developed by Muxika et al. (2007) to aggregate the measured species richness values, ShannonWiener diversity index and the AMBI index (Borja et al. 2000) into one global indicator. The following score system proposed by Muxika et al. (2007) for M-AMBI were used: high benthic alteration for 0 B M-AMBI \ 0.51, moderate alteration for 0.51 B M-AMBI B 0.71, low for 0.71 B M-AMBI \ 0.96 and no benthic alteration for M-AMBI C 0.96. Statistical analysis Univariate analyses were based on the toxicity test results and community descriptive parameters of the benthic fauna estimated, which were calculated for each replicate sample and summarised for each site. The results were expressed as the arithmetic mean ± standard deviations. The acute toxicity for the Microtox BSPT bioassay was analysed using a gamma model of MicrotoxOmni software (SDI Europe, Hampshire, UK). The survival data (A. brevicornis and P. lividus bioassays) were arcsine transformed for normality. A one-way analysis of variance (ANOVA) was used to test the significance of the physicochemical and benthic univariate measures and the effect caused by exposure to various types of sediment samplings compared to the reference or control samples and was followed by multiple comparisons of Dunett’s tests. Univariate statistical analyses of toxicity tests were performed using SPSS 15.0 (International Business Machines Corporation, Massachusetts, USA). The control acceptability criterion of the bioassay was set at C80 % for A. brevicornis survival in whole sediments and embryogenesis success of P. lividus larvae in elutriates (Cesar et al. 2004; Marıń et al. 2007). Univariate measures of benthic faunal data were performed with the computer software package PRIMER 6.1.6 (Plymouth Routines in Multivariate Ecological Research)

123

C. Silva et al.

(Clarke and Warwick, 2001). The statistical significance was set at P \ 0.05. For multivariate analyses, the species abundance was transformed with fourth root prior to performing a distance-based permutational multivariable analysis of variance (PERMANOVA). The statistical significance of differences in the benthic fauna abundance among sampling sites were analyzed with PERMANOVA based on Bray-Curtis similarity measures (Anderson et al. 2008). PERMANOVA software was used for testing the simultaneous response of fauna to one factor in a one-way ANOVA experimental design on the basis of any distance measure, using permutation methods (Anderson, 2001). For the one-way case, an exact P value was provided using unrestricted permutation of raw data. In addition, a posteriori pairwise comparisons were performed. When low unique values in the permutation distribution were available, asymptotical Monte Carlo P values (PMC) were used instead of permutational P values (PPERM). A 1-factor crossed design was used with sampling sites (5 levels: fixed) as factor and three replicates. The PERMANOVA analysis were performed using PRIMER 6.1.6. A multivariate analysis using a principal components analysis (PCA) extraction procedure was also performed to evaluate possible associations between sediment chemical data and the results of the benthic fauna univariate measures and toxicity bioassays. Microtox EC50 results after 30 min of exposure were used in the multivariate analysis. The PCA analysis was performed using SPSS 15.0.

Results Physicochemical characteristics The summarised results for the physicochemical characteristics of adjacent water, sediments and interstitial water by sampling site are shown in Table 2. No significant spatial differences were observed in water SA and T between the sampling sites. All the sampling sites were characterised by muddy bottom sediments. In general, a spatial gradient of contamination was observed from the aquaculture effluent (RSP1) to the reference site (RSP5). RSP1, which was the site closest to the aquaculture effluent, was characterised by lower levels of Eh, pH and DO values and higher values of TOM, Cu, Ni, Pb and Zn. By contrast, the reference site (RSP5) exhibited higher Eh, pH and DO and lower levels of metals (Cu, Ni and Pb) and organic matter. Metal concentrations were compared with sediment quality values (SQV) for the Gulf of Cadiz (Choueri et al. 2009) (Table 3). The SQV used two thresholds (SQV-low and SQV-high) that were designed for highly polluted ecosystems to avoid false negatives. Sediments from RSP1


1003

Table 2 Average values and standard deviations (in brackets) of each physicochemical variable of water (T and SA), sediment (Eh, pH, TOM and metal concentrations) and interstitial water (DO) at the different sampling sites in the Rıó San Pedro creek Variable

RSP1

T (°C)

RSP2

22.68 (0.30)

24.02 (0.10)

RSP3 21.51 (0.40)

RSP4 24.35 (0.20)

RSP5 22.54 (0.30)

SA (psu)

39.73 (0.01)

39.83 (0.02)

39.07 (0.05)

39.77 (0.02)

38.54 (0.04)

Eh (mV)

-145.8 (8.9)*

-126.1 (5.2)*

-103.4 (4.5)*

-101.3 (9.1)*

-65.7 (3.8)*

pH

7.084 (0.041)*

7.218 (0.022)*

7.254 (0.015)*

7.326 (0.036)*

7.420 (0.045)*

DO (mg L-1)

0.082 (0.011)*

0.055 (0.007)*

1.191 (0.018)*

2.154 (0.023)*

2.402 (0.030)*

17.5 (0.1)*

16.6 (0.1)*

12.2 (0.2)*

4.5 (0.1)*

3.8 (0.1)*

Cd (lg g ) Cu (lg g-1)

0.70 (0.04)* 151.0 (7.0)*

0.43 (0.03)* 31.6 (0.1)*

1.13 (0.05) 26.3 (2.3)*

1.12 (0.02) 15.1 (0.5)

1.03 (0.08) 16.5 (1.0)

Ni (lg g-1)

45.0 (2.0)*

37.4 (0.2)*

29.7 (1.1)*

19.0 (0.2)

18.3 (1.0)

Pb (lg g-1)

89.7 (6.5)*

23.8 (0.6)*

26.4 (0.3)*

15.4 (0.3)

15.2 (0.2)

Zn (lg g-1)

166.0 (1.0)*

92.3 (6.3)*

82.0 (6.0)*

50.4 (2.0)

56.2 (5.0)

TOM (%) -1

Asterisks (*) indicate significant differences (P \ 0.05) compared with the reference site RSP5

Table 3 Sediment quality guidelines proposed by Choueri et al. (2009) for the Gulf of Ca´diz Metals

Sediment quality values Not polluted

Moderately polluted

Highly polluted

Cd (lg g-1)

B0.65

[0.65 and \1.2

C1.2

Cu (lg g-1)

B20.8

[20.8 and \169.0

C169.0

Ni (lg g-1) Pb (lg g-1)

B8.9 B21.6

[8.9 and \42.3 [21.6 and \99.2

C42.3 C99.2

Zn (lg g-1)

B138.0

[138.0 and \360.0

C360.0

were toxic or polluted. The Ni in the RSP1 was higher than the SQV-high values. Sediments from other sites were classified as moderately polluted. The RSP5 and RSP4 sites were the least polluted. Bioassays Table 4 shows the EC50 results obtained by the Microtox test with the sediments collected at the RSP sampling sites. Significant differences (P \ 0.05) in the EC50 values between the sampling sites and the reference site (RSP5) were observed, for all time intervals (5, 15 and 30 min of exposure). In all cases, RSP5 presented the highest EC50 values, while the lowest were accounted at RSP1, which is located the closest to the fish farm effluent (Table 4). The EC50 results after 30 min of exposure at the sampling sites (RSP1 to RSP4) were under the limits assumed for sediment toxicity: 1,000 mg L-1 (Canadian Standards in Environment Canada 2002) and 750 mg L-1 (Spanish standards in DelValls et al. 2004; Casado-Martıńez et al. 2006a). The mean EC50 value for the sediments of RSP5 site was 2,038.29 mg L-1, which exceeded the limit for

sediment acute toxicity. The highest toxicity to V. fischeri, expressed as the lowest EC50 after 30 min, were observed at RSP1, RSP2 and RSP3, which are the sites closest to the effluent. The results obtained from the 10-day bioassay with A. brevicornis are shown in Table 4. Elevated toxicity was measured for sediments RSP1 to RSP4. Stronger toxicity effects were observed at sites RSP1 and RSP2 sediments, which were closest to the fish farm effluent, with mean survival values ranging from 40 to 45 %. There were not significant differences (P \ 0.05) between the clean site (RSP5, 95 % survival) and the control site (CA, 100 % survival). The percentage success of normally developed P. lividus larvae at the sampling stations in the RSP creek is shown in Table 4. A decreased number of normal P. lividus larvae were observed in the polluted sites located closer to the fish farm effluent. The mean values of success ranged from 45 % (RSP2) to 82 % (RSP5), with 90 % success in the control sample CSW. The results indicate the presence of adverse effects (percentage of success \80 %) in sediment elutriates from stations RSP1 to RSP4, which showed significant differences (P \ 0.05) compared to the control sample CSW. The toxicity values obtained from the integration of the results of the different bioassays (Microtox EC50, reduction in A. brevicornis survival and in P. lividus larval developmental success) for each sampling site using the scores in Table 1 are shown in Table 5. RSP5 had the lowest toxic score (TI = 1), so it was considered ‘‘non-toxic’’. By contrast, the highest toxicity index (TI = 6) corresponded to sediments from sampling sites RSP1, RSP2 and RSP3. The RSP4 sampling site showed a high-moderate toxicity (TI = 5).

123

1004

C. Silva et al.

Table 4 Average values and standard deviations (in brackets) of toxicity tests (A. brevicornis and P. lividus: % survival; V. fischeri: EC50 effective concentration in mg sediment L-1 aqueous extract) and benthic fauna alteration (S: total species; N: number of Toxicity tests V. fischeri 5 min

V. fischeri 30 min

individuals per 0.025 m-2; H’: Shannon diversity index) results for sediments from the different sampling sites in the Rıó San Pedro creek A. brevicornis

P. lividus

Benthic fauna alteration

V. fischeri 15 min

S

N

H’

RSP1

422.4 (165.3)*

312.4 (146.3)*

340.0 (59.0)*

40 (8.7)*

50 (5.1)*

3.3 (0.6)*

27.3 (3.5)*

0.8 (0.2)*

RSP2

5,444.5 (1,345.6)*

1,533.7 (346.6)*

419.4 (231.7)*

45 (8.7)*

45 (8.1)*

4.7 (1.5)*

42.3 (7.8)*

1.0 (0.1)*

RSP3

2,628.3 (599.8)*

765.4 (152.3)*

299.1 (103.3)*

68 (2.9)*

54 (1.7)*

9.0 (2.6)

50.7 (8.6)*

1.4 (0.3)*

RSP4

3,320.0 (1,496.4)*

1,083.0 (279.4)*

702.0 (151.5)*

65 (5)*

68 (2.9)*

8.7 (2.5)

114.0 (16.1)*

1.8 (0.1)

RSP5

23,226.3 (2,200.3)

3,707.0 (405.9)

2,038.3 (181.8)

95 (5)

82 (4.7)

12.7 (1.2)

139.0 (11.4)

2.0 (0.1)

CA

100 (0)

CSW

90 (1.8)

RSP1–RSP5 Rıó San Pedro sampling sites, RSP5 reference site, CA control site, CSW control sample of clean sea water. Asterisks (*) indicate significant differences (P \ 0.05) compared with the control or reference site RSP5 Table 5 Integrated toxicity and benthic alteration indices for sediments from the different sampling sites in the Rıó San Pedro creek Integrated index

RSP1

RSP2

RSP3

RSP4

RSP5

Toxicity index (TI)

6

6

6

5

1

M-AMBI

0.5 (high)

0.59 (moderate)

0.74 (Low)

0.88 (Low)

0.98 (no alteration)

RSP1–RSP5, Rıó San Pedro sampling sites

Benthic community Twenty-six species were identified in a total of 1,120 individuals collected in the study area, belonging to mollusks (9 species), polychaetes (5 species), crustaceans (5 species), nematodes (4 species) and oligochaetes (3 species). All the benthic community univariate indices followed very similar trends at the different sites, characterised by a spatial gradient with higher values in the reference or clean site and lower in the polluted sites close to the fish farm (Table 4). The total species (S) average value ranged from 12.7 (reference site RSP5) to 3.3 (RSP1), and a spatial variability or gradient was observed with an increase in S from the pollution effluent to the reference site (Table 4). Significant (P \ 0.05) differences were observed between sampling sites (RSP1 and RSP2) close to the effluent and the reference site RSP5. Significant (P \ 0.05) differences among the sampling sites were also observed for abundance N, with a spatial gradient characterised by greatest values in the reference site and lower values in the sites close to the fish farm effluent (Table 4). The biodiversity index H’ showed significant (P \ 0.05) differences between sampling sites and followed similar spatial fluctuations (Table 4). The M-AMBI benthic alteration index ranged from 0.98 (no alteration) in the clean area to 0.5 (high alteration) in the area impacted by fish farm discharges (Table 5). A spatial gradient is observed with lower values of the

123

M-AMBI indicator in the upper part of the RSP, which is associated with the area contaminated by the aquaculture discharges. PERMANOVA revealed highly significant (PPERM = 0.001) differences in benthic fauna composition among sampling sites (Table 6). Pairwise tests showed that a highly significant (PMC = 0.001) difference in fauna occurred between the control site (RSP5) and the site nearest to the aquaculture effluent (RSP1). Significant (PMC \ 0.05) difference occurred among the other sites except for RSP2 and RSP3, which have no statistically dissimilarities. Associations between environmental physicochemical characteristics, toxicity and benthic community The relations between the site pattern in the acute toxicity tests, the benthic community alteration and the environmental variables are presented in the PCA plot (Fig. 2; Table 7). The first three principal components (PC) explained 94.87 % of the total variability, with the first principal component (PC1) representing 41.81 % of the total variability, the second (PC2) representing 32.38 % and the third (PC3) representing 20.68 %. The criteria for considering a variable as being associated with a particular factor was defined as its having a loading of 0.4 or higher, which approximates Comreys’ cut-off (Comreys 1973) of 0.6 or higher for a reasonable association between an


1005

Table 6 Results of PERMANOVA based on Bray-Curtis dissimilarities (%) to test for differences in benthic fauna abundance at each sampling site (PMC—values are shown in parenthesis) Source Site

df 4

Residuals

10

Total

14

MS

F

PPERM

2,987.40

13.85

0.001

215.75 Within sites

Among sites

RSP1 = 13.51

RSP1 versus RSP2 = 33.28 (0.019)

RSP2 versus RSP4 = 46.23 (0.013)

RSP2 = 25.54

RSP1 versus RSP3 = 44.04 (0.009)

RSP2 versus RSP5 = 54.24 (0.003)

RSP3 = 21.87 RSP4 = 25.87

RSP1 versus RSP4 = 59.54 (0.003) RSP1 versus RSP5 = 63.86 (0.001)

RSP3 versus RSP4 = 43.43 (0.016) RSP3 versus RSP5 = 50.35 (0.003)

RSP5 = 9.26

RSP2 versus RSP3 = 31.82 (0.076)

RSP4 versus RSP5 = 32.51 (0.027)

original variable and a factor and takes into account discontinuities in the magnitudes of the loadings of the original variables (DelValls et al. 1998b; Martıń-Dıáz et al. 2008). PC1 clearly separated the reference clean site RSP5 at one extreme from other sampling stations (RSP1, RSP2 and RSP3) at the opposite extreme. In the axis PC1, the sampling site RSP5 is closely associated with higher values of Microtox EC50, success of normally developed P. lividus larvae, A. brevicornis survival, benthic fauna abundance, number of species, biodiversity, distance to effluent (D), Eh, pH and DO, and lower Ni concentration and TOM content. These findings indicated an increase in organic matter, acidification, reduction of sediments, metal contents and anoxic conditions of sediments from the right to the left of the diagram (Fig. 2), mainly evidenced at sampling sites RSP1, RSP2, RSP3 and RSP4, which were closer to the aquaculture effluent. PC2 separates the RSP1 site, which is the closest to the aquaculture effluent, from all other sampling stations; this station is associated with higher metal contents (Cu, Ni, Pb and Zn) and TOM content and lower Eh, pH, DO, A. brevicornis survival, number of benthic species, fauna abundance and Shannon biodiversity. PC3 separates RSP2 from all other sampling stations and is associated with higher TOM and Ni contents as well as lower Cd, DO, success of normally developed P. lividus larvae and the univariate benthic alteration indicators (S, N and H’).

Discussion The effects of discharged wastes from fish aquaculture activities on sediment quality of some European coastal areas are well documented (Mantzavrakos et al. 2007; Pusceddu et al. 2007; Vezzulli et al. 2008; Holmer 2010). To improve the methods used for fish aquaculture impact monitoring, a biological study of the potential acute toxicological effects of the marine sediments that are impacted by LBMFF effluents was carried out by applying a battery

of bioassays and benthic community measurements. This study permitted an examination of the associations between environmental conditions of marine sediments and the biological responses on population dynamics and at different trophic levels: bacteria (V. fischeri), amphipods (A. brevicornis) and sea urchins (P. lividus). Higher contamination levels were observed at sites near the aquaculture effluent, and exposure to organic enrichment and metals in the sediments from these sites caused adverse effects in all toxicity tests and in the benthic community compared to the reference site RSP5. It was demonstrated that discharges from LBMFFs could lead to acute toxicity in exposed organisms from different trophic levels. In this study, a benthic organic enrichment surrounding a fish farm effluent following a pollution gradient was evidenced and accompanied by a change in sediment physicochemical characteristics. A spatial gradient was observed in TOM and metal contents (Cu, Ni, Pb and Zn), with higher values in the upper part of the creek in the sampling sites closest to the LBMFF effluent. Lower values of Eh and pH were observed in the areas surrounding the fish farm, with significant (P \ 0.01) differences compared to the reference site RSP5. The Eh values indicate that the trend towards reduced conditions is due to the effects of the fish farm effluent on the sediment. Eh is a qualitative metric of the intensity of reduction conditions (SEPA 2005) that could reflect organic enrichment, but low Eh could also be the result of natural physical processes, especially in soft sediments (Brooks et al. 2002). Therefore, its usefulness in assessing the effects of fish farms is limited if it is not accompanied by a deep knowledge of the study area. To correctly measure Eh in sediments, the sampling method must ensure that the vertical profile is not altered, which was guaranteed by the in situ sampling in the intertidal muddy sediment during neap tide. Organic enrichment of the seabed involves sediment anoxia, increased nutrient levels, deterioration of the optical properties of seawater and impoverished sediment quality due to organic matter sedimentation (Holmer et al. 2005).

123

1006

C. Silva et al.

(a) 1

Table 7 Rotated loading (coefficient) matrix provided by the multivariate analysis in order to define factors or principal components PC1, PC2 and PC3 pH H'

Eh S N DO D Amph Urch EC50

PC2 (32.38 %)

0.5 RSP2-1 RSP4-1 RSP2-2 RSP2-3 RSP4-2 RSP3-3 RSP4-3 RSP3-2 RSP3-1

0

Cd

RSP5-1 RSP5-2 RSP5-3

TOM

-0.5

Ni RSP1-3 RSP1-1 Zn Pb

-1 -1

RSP1-2 Cu

-0.5

0

(b) 1

0.5

1

Cd

DO RSP4-2

0.5

PC3 (20.68 %)

RSP3-2

H'

RSP4-1

RSP4-3 RSP3-1 RSP3-3

Amph EC50

RSP5-1

RSP1-1

Pb

RSP5-3

RSP1-3 Zn

-0.5

Ni TOM

RSP2-2 RSP2-1

-1 -1

-0.5

RSP2-3

0

0.5

1

PC1 (41.81 %) Site-replicate

Environment

Bioassay

Fauna

Fig. 2 Principal Component Analysis (PCA) of the environmental, acute toxicity tests and benthic fauna parameters from the five sampling sites within the Rio San Pedro creek during May 2011: a PC1 versus PC2, b PC1 versus PC3. Urch, Sea urchin toxcicity test; Amph, Amphipod toxicity test; EC50, Microtox toxicity test

The fact that a high ([10 %) TOM content (Long and Morgan 1990) is associated with sediment anoxia is supported by the negative association with the DO concentration as estimated in the PCA analysis of the present study. Metal contamination that is derived from anaerobic mineralisation of organic matter is also supported by the higher concentrations of Cu, Ni, Pb, Zn and TOM and lower levels of DO observed in the sites closest to the fish farm effluent. The spatial pollution gradient observed in

123

PC2 32.38

PC3 20.68

D

0.862

0.327

0.366

Eh

0.685

0.541

0.377

pH

0.619

0.681

0.305

DO

0.606

0.431

0.661

TOM

20.613

20.468

20.598

Cu

-0.202

20.959

-0.173

Zn

-0.345

20.873

-0.324

Cd

0.223

0.183

0.936

Ni

20.537

20.639

20.535

Pb

-0.240

20.957

-0.138

V. fischeri EC50

0.969

0.197

0.038

A. brevicornis

0.864

0.313

0.255

P. lividus

0.878

0.184

0.408

S N

0.652 0.727

0.453 0.439

0.428 0.416

H’

0.622

0.550

0.492

D

pH RSP5-2

RSP1-2

0

PC1 41.81

Each factor is defined by variables whose coefficient is C0.4 (in bold)

N Urch

S Eh

% variance

this study is consistent with previous results on the longitudinal distribution of various water quality parameters in the RSP creek (Tovar et al. 2000a, b). The higher TOM contents and lower values of pH, Eh and DO observed in this study in the sediments located closest to the aquaculture effluent are consistent with the increased contents of nutrients, particulate organic matter and suspended solids as well as decreased pH observed in the water column by Tovar et al. (2000a, b). These chemical conditions were due to the high ammonium concentration and the acidic characteristics of faeces and manufactured fish feeds Tovar et al. (2000a). Tovar et al. (2000a) identified the existence of two different zones within the inlet according to water quality. The first zone has a length of approximately 8 km and is closer to the mouth; there are no fish farms but good renovation of water as controlled by the tides. The second zone is affected by semi-intensive fish culture, and its effluents are considered the main source of nutrients, organic matter and suspended solids received by the ecosystem. The spatial pattern and higher metal contents (Cu, Pb and Zn) obtained in this study are consistent with the results of Mendiguchıá et al. (2006), who observed a significant enrichment of metals and organic matter in sediments that originated from LBMFF effluents. Trace metal pollution in coastal sediments in the Bay of Cadiz has been documented in previous field studies (Arau´jo et al. 2009; Blasco et al. 2010; Mendiguchıá et al. 2006). In general, sediments from the RSP1 site had the highest metal contamination levels, which were classified as toxic when


compared to the SQVs. The metal concentrations present in the tested RSP sediments were compared to the values obtained at the same locations in previous research (Blasco et al. 2010; Mendiguchıá et al. 2006). From these results, it was inferred that the metals concentrations are slightly higher for the RSP1 site, and are within the expected order of magnitude for the RSP2 through RSP5 sites. Metal contamination from aquaculture discharge and anaerobic mineralisation of organic matter is supported by the significant positive correlations between TOM and Cu, Ni, Pb and Zn. Remarkably, the aqua regia extraction method (ISO11466 1995) is a strong acid digestion that liberated a higher quantity of metals than others methods (e.g. partial and sequential extractions), and must be carefully taken into account in future assessments. However, this extraction method was previously used to assess metal concentration in the study area (Blasco et al. 2010; Mendiguchıá et al. 2006). In this study, the V. fischeri, A. brevicornis and P. lividus toxicity tests were sensitive bioassays to assess the potential effects of LBMFF effluents in marine sediments. The results show a clear association between chemical contamination and toxicity because the most contaminated site (RSP1) was the most toxic site, and the less contaminated or reference site (RSP5) was not toxic. The Microtox test has been successfully applied to test the toxicity of chemical compounds that are used in fish farming, such as antibiotics (Lalumera et al. 2004; Isidori et al. 2005) and organics (Hernando et al. 2007). According to the criteria proposed in Spain (EC50 \ 750 mg L-1 d.w.) (MoralesCaselles et al. 2007), the sampling stations RSP1 to RSP4 were toxic, and the reference site (RSP5) was non-toxic to the Microtox assay (EC50 [ 750 mg L-1 d.w.; 30 min exposure). An evident spatial pattern in sediment toxicity and metal concentrations was observed at the study site. Microtox EC50 values are positively associated (PC1) with the distance to the effluent, Eh, pH and DO and negatively correlated with TOM content and Ni concentration. Previous studies have shown that the amphipod species A. brevicornis is a sensitive organism to assess the toxicity of contaminated sediments (Riba et al. 2003; Casado-Martıńez et al. 2006b; Alvarez-Guerra et al. 2009; RamosGo´mez et al. 2009). According to the criteria used (a greater than 20 % reduction in A. brevicornis survival), the sampling stations RSP1 to RSP4 were toxic, and RSP5 was non-toxic in the A. brevicornis survival bioassay. The spatial contamination in A. brevicornis survival was associated with higher toxicity and sediments chemical concentrations. Higher A. brevicornis survival was associated (PC1) with greater distance from the effluent, high Eh, pH and DO and lower Ni and TOM contents. Additionally, higher A. brevicornis survival was related (PC2) to high Eh, pH and DO and low TOM and metal (Cu, Pb, Ni, Zn)

1007

contents. The P. lividus bioassay, in which the percentage success of normally developed larvae is measured, is a sensitive and accurate method of determining acute toxic effects of sediments affected by fish farm effluents. These results are consistent with those of Carballeira et al. (2011a), who developed standardised bioassay tests to assess the toxicity of water effluents from eight LBMFFs located in the north-west coast of Spain. Toxic effects of fish farm sediments were previously assessed using P. lividus larval bioassays (Marıń et al. 2007). During high production periods (spring and summer), a toxic response was observed by the P. lividus bioassay (Marıń et al. 2007). The obtained results indicate adverse effects (percentage success \80 %) in sediment elutriate toxicity tests for stations RSP1 to RSP4. Lower P. lividus values are associated with decreased distance to the effluent, Eh, pH and DO, and higher metal (Ni) concentrations and organic matter contents. The toxicity values derived from the three toxicity tests confirmed the high sensitivity of the bioassays in assessing the toxicity of sediments impacted by LBMFF effluents. The index correctly integrated the potential toxicity of sediments, particularly from high (RSP1 to RSP3) and low (RSP5) toxic sampling sites. The results of toxicity tests have previously been integrated by means of a scoring system and statistical summarisation methods (Chapman et al. 2002; Alvarez-Guerra et al. 2009). Apart from the use of chemical analysis and laboratory bioassays presented in this study for sediment quality assessment, the inclusion of biomarkers, in situ bioassays, bioaccumulation and biomagnification (Chapman and Hollert 2006; Chapman 2007), could also be required, depending on the specific case under study. The V. fischeri, A. brevicornis and P. lividus toxicity tests were also complementary of each other in terms of test matrices employed. Whole sediment was assayed with the Microtox BSPT and amphipods tests to provide information on particulate and adsorbed contaminants effects. One major advantage of assaying whole sediment as opposed to aqueous extracts is that the organism is in direct contact with the sample particles during the incubation (bacteria) and exposure (amphipod) period. However, the lack of standardised methods and appropriate species for whole sediment tests make routine bioassays of this matrix problematic (Macken et al. 2008). To avoid these problems the use of complementary bioassays with aqueous extracts (e.g. elutriates) have been developed. Sediment elutriation provide information on the water-soluble contaminants and enables to assess the potential effects in sensitive biological components of the water column, including gametes, embryos and larvae (Beiras 2002; Carballeira et al. 2012). Therefore the most suitable method to assess the potential effects of LBMFF effluents in marine sediments would be

123

1008

one which incorporates both solid-phase and elutriates testing in the assessment process. The integrated results of the three bioassays (V. fischeri, A. brevicornis and P. lividus) concurred in classifying the sediment samplings of stations RSP1 to RSP4 as toxic and RSP5 as non-toxic. Effects were detected with all the bioassays in this study demonstrating that these tools are sensitive standard tests to monitor sediment toxicity. The spatial variations of bioassay results and associations with sediment contamination found in this study highlight the need to use a multi-trophic battery of toxicity tests to assess sediment quality. The obtained results suggest that the toxicity responses of V. fischeri, A. brevicornis and P. lividus could be influenced by a combination of metal contaminants, organic enrichment and anoxic and reducing redox conditions of sediments. The validity of the proposed approach for assessing the impact of LBMFF effluents on marine sediment quality, which include three multi-trophic acute bioassays, has been demonstrated. The effects of stress on population dynamics include the following: an increase in production linked to eutrophication, a reduction in diversity and a decrease in the number of species with increasing number of individuals (Pearson and Black, 2001). Benthic faunal studies have been used to assess in situ alterations in residential community structure in relation to pollution-induced changes derived from aquaculture activities (Pusceddu et al. 2007; Tomassetti and Porrello, 2005; Vezzulli et al. 2008). In the RSP, this benthic community analysis has not been difficult due to the existence of an appropriate reference site (RSP5) at the mouth of the creek. This site is located in an area that does not receive any recognised pollution inputs and has good water circulation. It is clear that the benthic community has been affected by the contamination of the fish farm effluent located in the upper part of the creek. Sediment alteration in the more contaminated sites (RSP1, RSP2 and RSP3) was due to an increase in fish farm waste and low water renovation, allowing accumulation of organic matter, oxidative conditions and anoxia in the area of the sea floor. In situ alterations were reflected by changes in the benthic community that followed a spatial gradient of contamination from the aquaculture effluent to the reference site. Benthic community at the more contaminated sites with respect to the control site, show perturbations caused by organic enrichment: fewer species in sampling sites near the effluent; lower number of individuals and diversity indexes; PCA plots showing clear separation between clean (RSP5) and more contaminated (RSP1, RSP2 and RSP3) sites. The analysis performed across spatial scale provides additional insights into how benthic fauna respond at the different sites following a spatial gradient of contamination. The PERMANOVA analysis revealed highly significant differences in benthic fauna structure among the

123

C. Silva et al.

reference site and the site close to the aquaculture effluent, which suggests that the effects of fish farming vary across the spatial gradient. In the present study the most common and abundant species at the reference site, which dominate the intertidal benthic community in muddy sediments, were similar as those reported in previous studies for similar unpolluted ecosystems in the Gulf of Ca´diz and Portugal, including mollusks (Scrobicularia plana and Hydrobia ulvae), oligochaetes and polychaetes (Paradoneis lyra, Nereis diversicolor) (DelValls et al. 1998b; França et al. 2009; Rodrigues et al. 2006). The predominant benthic fauna found at the most contaminated sites were isopods Cyathura carinata, polychaetes (P. lyra, N. diversicolor) and nematodes. These species pointed out the presence of a similar community to the one described as ‘Reduced Macoma’ by Thorson (1957) and characteristic of estuary sites with salinity stress. However, oligochaetes, which are a group of taxa considered more tolerant to pollution, were not found in the more contaminated sites (DelValls et al. 1998b). Additionally, the soft bottom benthic community affected by fish farm effluents shows a prevalence of small species (Macleod et al. 2004; Heilskov et al. 2006) in line with the Pearson and Rosemberg (1978) model. This model predicts a reduced mean body size, a shallower distribution and an impoverished functional community structure with increasing organic load (Heilskov et al. 2006). The integrated results of the benthic fauna using the M-AMBI index classified the sediment sites of RSP1 as high, RSP2 as moderate and RSP3 and RSP4 as low alteration, while the reference sites is not altered. The obtained results suggest that the benthic fauna community structure could be influenced by a combination of metal contaminants, organic enrichment and anoxic and reducing redox conditions of sediments. It has been demonstrated that discharges from fish farms may lead to a stress on benthic fauna population dynamic including a loss of biodiversity, number of species and abundances. Several studies have highlighted the limitations of traditional monitoring of physicochemical parameters and support the use of new monitoring tools to assess the impact of fish farm effluents (Marıń et al. 2007; Carballeira et al. 2011a). The information obtained from physicochemical analyses is insufficient because emerging pollutants are not taken into account, and the data do not indicate the potential effects on ecological processes (Chapman 2007) because they do not reflect the bioavailability of contaminants (Hernando et al. 2007). Consequently, the set of bioassay tools applied in this study satisfy the requirements for their consideration as useful techniques for sediment quality and risk assessment in areas that are impacted by LBMFF effluents. The use of a battery of laboratory bioassays to evaluate the potential toxicity of fish farm effluents in marine sediments,


combined with the study of the physicochemical and benthic community characteristics in the area, provide accurate information about the spatial variability of pollutants and potential biological effects at different trophic levels.

Conclusion Rio San Pedro presented a pollution gradient from the fish farm effluent to the reference site characterised by hypoxia/ anoxia, reduced potential, acidic conditions, high organic enrichment and metals (Cu, Pb, Ni, Zn) in sediments at the most contaminated sites. The results of the three bioassays (V. fischeri, A. brevicornis and P. lividus) revealed a decrease in Microtox EC50, A. brevicornis survival, P. lividus larval development success and biodiversity indicators (S, N, d H’) of the benthic community in intertidal sediments closer to the fish farm. This decrease was produced by the effects of aquaculture discharged effluents in sediments, which is a strong indication that severe acute effects exist in this shallow tidal creek. The multivariate analysis allowed the identification of chemical sediment conditions along the creek, causing ecotoxicological changes and thus potential dangers for benthic organisms. It has been demonstrated that effluents from fish aquaculture activities may induce acute toxicity in soft-sediment species, which may lead to an alteration of the biodiversity of the exposed organisms. The three bioassays used in this research proved to be successful due to being easy, rapid and sensitive, and organisms presented favourable characteristics for their use in toxicity test, and they can be used as standard organism to evaluate the potential effects of fish farm effluents in marine sediments. Rich data sets will improve confidence in the integrated ecotoxicological assessment of marine sediments affected by fish farm effluents, but even in data-poor contexts, this kind of approach can assist in making decisions based on physicochemical, bioassays and benthic community measurements. This integrative approach can be applied in data-poor countries with high aquaculture sector growth to improve environmental management and ecological risk assessment Acknowledgments C. Silva is grateful to the Sistema Bicentenario BECAS CHILE from the Chilean Government and the international grant from Bank Santander/UNESCO Chair UNITWIN/WiCop for funding this work. The work described was also supported partially by grants from the Spanish Ministry of Science and Innovation (CTM2011-28437-C02-02). The authors are grateful to the anonymous reviewers for their valuable comments and suggestions to improve the quality of the paper.

1009

References Alvarez-Guerra M, Viguri JR, Voulvoulis N (2009) A multicriteriabased methodology for site prioritisation in sediment management. Environ Int 35:920–930 Anderson MJ (2001) A new method for non-parametric multivariate analysis of variance. Aust J Ecol 26:32–46 Anderson MJ, Gorley RN, Clarke KR (2008) PERMANOVA ? for PRIMER: guide to software and statistical methods. PRIMER-E, Plymouth Antunes P, Gil O (2004) PCB and DDT contamination in cultivated and wild sea bass from Ria de Aveiro, Portugal. Chemosphere 54(10):1503–1507 Arau´jo CVM, Diz FR, Laiz I, Lubiań LM, Blasco J, Moreno- Garrido I (2009) Sediment integrative assessment of the Bay of Ca´diz (Spain): an ecotoxicological and chemical approach. Environ Int 35(6):831–841 Azur Environmental (1998) Microtox acute toxicity basic test procedures. Azur Environmental, Carlsbad, CA Beiras R (2002) Comparison of methods to obtain a liquid phase in marine sediment toxicity bioassays with Paracentrotus lividus sea urchin embryos. Arch Environ Contam Toxicol 42:23–28 Beiras R, Fernańdez N, Bellas J, Besada V, Gonza´lez-Quijano A, Nunes T (2003) Integrative assessment of marine pollution in Galician estuaries using sediment chemistry, mussel bioaccumulation, and embryo-larval toxicity bioassays. Chemosphere 52:1209–1224 Blasco J, Gomes T, Garcıá-Barrera T, Rodriguez-Romero A, Gonza´lez-Rey M, Morań-Roldań F, Trombini C, Miotk M, Go´mez-Ariza JL, Bebianno M (2010) Trace metal concentrations in sediments from the southwest of the Iberian Peninsula. Sci Mar 74S1:99–106 Borja A, Franco J, Pe´rez V (2000) A marine biotic index to establish the ecological quality of soft-bottom benthos within European estuarine and coastal environments. Mar Poll Bull 40:1100–1114 Brooks KM, Mahnken C, Nash C (2002) Environmental effects associated with marine net pen waste with emphasis on salmon farming in the Pacific Northwest. In: Stickney RR, McVey JP (eds) Responsible marine aquaculture. World aquaculture society. CABI Publishing, Wallingford, p 391 Cabello FC (2006) Heavy use of prophylactic antibiotics in aquaculture: a growing problem for human and animal health and for the environment. Environ Microbiol 8(7):1137–1144 Carballeira C, De Orte MR, VianaI IG, Carballeira A (2011a) Implementation of a minimal biological test set for assessment of ecotoxic effect of effluents from land-based fish farms. Ecotox Environ Safe 78:148–161 Carballeira C, Martıń-Dıáz L, DelValls TA (2011b) Influence of salinity on fertilization and larval development toxicity tests with two species of sea urchin. Mar Environ Res 72:196–203 Carballeira C, Ramos-Go´mez J, Martıń-Dıáz L, DelValls TA (2012) Identification of specific malformations of sea urchin larvae for toxicity assessment: application to marine pisciculture effluents. Mar Environ Res 77:12–22 Casado-Martıńez MC, Campisi T, Dıáz A, Lo Re R, Obispo R, Postma JF, Riba I, Sneekes AC, Buceta JL, DelValls TA (2006a) Interlaboratory assessment of marine bioassays to evaluate the environmental quality of coastal sediments in Spain. II. Bioluminescence inhibition test for rapid sediment toxicity assessment. Cienc Mar 32:129–138 Casado-Martıńez MC, Beiras R, Belzunce MJ, Gonza´lez-Castromil MA, Marin-Guirao L, Postma JF, Riba I, DelValls TA (2006b) Interlaboratory assessment of marine bioassays to evaluate the environmental quality of coastal sediments in Spain. IV. Whole

123

1010 sediment toxicity test using crustacean amphipods. Cienc Mar 32:149–157 Casado-Martıńez MC, Fernańdez N, Lloret J, Marıń A, Martıńez´ lvarez L, DelValls TA Go´mez C, Riba I, Beiras R, Saco-A (2006c) Interlaboratory assessment of marine bioassays to evaluate the environmental quality of coastal sediments in Spain. III. Bioassay using embryos of the sea urchin Paracentrotus lividus. Cienc Mar 32:139–147 Cesar A, Marıń A, Marıń-Guirao L, Vita R (2004) Amphipod and sea urchin tests to assess the toxicity of Mediterranean sediments: the case of Portman Bay. Sci Mar 68:205–213 Cesar A, Marıń A, Marin-Guirao L, Vita R, Lloret J, DelValls TA (2009) Integrative ecotoxicological assessment of sediment in Portmań Bay (southeast Spain). Ecotox Environ Safe 72:1832–1841 Chapman PM (1996) Presentation and interpretation of sediment quality triad data. Ecotoxicology 5:327–339 Chapman PM (2007) Determining when contamination is pollution — weight of evidence determinations for sediments and effluents. Environ Int 33(4):492–501 Chapman PM, Hollert H (2006) Should the sediment quality triad become a tetrad, a pentad, or possibly even a hexad? J Soils Sediments 6(1):4–8 Chapman PM, McDonald BG, Lawrence GS (2002) Weight of evidence frameworks for sediment quality and other assessments. Hum Ecol Risk Assess 8:1489–1515 Choueri RB, Cesar A, Abessa DMS, Torres RJ, Morais RD, Riba I, Pereira CDS, Nascimento MRL, Mozeto A, DelValls TA (2009) Development of site-specific sediment quality guidelines for North and South Atlantic littoral zones: comparison against national and international sediment quality benchmarks. J Hazard Mat 170:320–331 Clarke KR, Warwick RM (2001) Change in marine communities: an approach to statistical analysis and interpretation, 2nd edn. PRIMER-E Ltd. Plymouth Marine Laboratory, Plymouth Comreys AL (1973) A first course in factor analysis. Academic, New York De la Paz M, Go´mez-Parra A, Forja JM (2008a) Variability of the partial pressure of CO2 on a daily-to-seasonal time scale in a shallow coastal system affected by intensive aquaculture activities (Bay of Cadiz, SW Iberian Peninsula). Mar Chem 110:195–204 De la Paz M, Go´mez-Parra A, Forja JM (2008b) Tidal-to-seasonal variability in the parameters of the carbonate system in a shallow tidal creek influenced by anthropogenic inputs, Rıó San Pedro (SW Iberian Peninsula). Cont Shelf Res 28:1394–1404 DelValls TA, Forja JM, Gonza´lez-Mazo E, Go´mez-Parra A (1998a) Determining contamination sources in marine sediments using multivariate analysis. Trends Anal Chem 17(4):181–192 DelValls TA, Conradi M, Garcıá-Adiego E, Forja JM, Go´mez-Parra A (1998b) Analysis of macrobenthic community structure in relation to different environmental sources of contamination in two littoral ecosystems from the Gulf of Ca´diz (SW Spain). Hydrobiologia 385:59–70 DelValls TA, Forja JM, Go´mez-Parra A (1999) An integrative study of environmental quality of sediments from the Bay of Cadiz. Ecotoxicol Environ Res 2:26–33 DelValls TA, Andres A, Belzunce MJ, Buceta JL, Casado-Martinez MC, Castro R, Riba I, Viguri JR, Blasco J (2004) Chemical and ecotoxicological guidelines for managing disposal of dredged material. Trends Anal Chem 23:819–828 Environment Canada (1992a) Biological test method: acute test for sediment toxicity using marine or estuarine amphipods. Environmental Protection Service, Ottawa, ON, Report EPS 1/RM/ 26, p 111

123

C. Silva et al. Environment Canada (1992b) Biological test method: fertilization assay using echinoids (sea urchins and sand dollars). Environmental Protection Service, Ottawa, ON, Report EPS 1/RM/27, p 97 Environment Canada (2002) Biological test method: reference method for determining the toxicity of sediment using luminescent bacteria in a solid-phase test. Environmental Protection Service, Ottawa, ON, Report EPS 1/RM/42, p 83 Ferroń S, Ortega T, Forja JM (2009) Benthic fluxes in a tidal salt marsh creek affected by fish farm activities: Rıó San Pedro (Bay of Cadiz, SW Spain). Mar Chem 113:50–62 França S, Vinagre C, Pardal MA, Cabral HN (2009) Spatial and temporal patterns of benthic invertebrates in the Tagus estuary (Portugal): comparison between subtidal and intertidal mudflat. Sci Mar 73:307–318 Heilskov AC, Alperin M, Holmer M (2006) Benthic fauna bioirrigation effects on nutrient regeneration in fish farm sediments. J Exp Mar Biol Ecol 339:204–225 Hernando MD, De Vettori S, Martıńez Bueno MJ, Fernańdez-Alba AR (2007) Toxicity evaluation with Vibrio fischeri test of organic chemicals used in aquaculture. Chemosphere 68:724–730 Holmer M (2010) Environmental issues of fish farming in offshore waters: perspectives, concerns and research needs. Aquac Environ Interact 1:57–70 Holmer M, Wildish D, Hargrave BT (2005) Organic enrichment from marine finfish aquaculture and effects on sediment biogeochemical processes. In: Hargrave BT (ed) Environmental effects of marine finfish aquaculture. Springer, Berlin, pp 181–206 Isidori M, Lavorgna M, Nardelli A, Pascarella L, Parrella A (2005) Toxic and genotoxic evaluation of six antibiotics on non-target organisms. Sci Total Environ 346:87–98 ISO11466 (1995) Soil quality—extraction of trace elements soluble in aqua regia. International Organization for Standardization, Switzerland Kalantzi I, Karakassis I (2006) Benthic impacts of fish farming: metaanalysis of community and geochemical data. Mar Pollut Bull 52:479–483 Karakassis I, Tsapakis M, Smith CJ, Rumohr H (2002) Fish farming impacts in the Mediterranean studied through sediment profiling imagery. Mar Ecol Prog Series 227:125–133 Lalumera GM, Calamari D, Galli P, Castiglioni S, Crosa G, Fanelli R (2004) Preliminary investigation on the environmental occurrence and effects of antibiotics used in aquaculture in Italy. Chemosphere 54:661–668 Long ER, Morgan LG (1990) The potential for biological effects of sediment-sorbed contaminants tested in the National Status and Trends Program. National Oceanic and Atmospheric Administration Technical Memorandum NOS OMA 52, USA, p 175 Macken A, Giltrap M, Foley B, McGovern E, McHugh B, Davoren M (2008) An integrated approach to the toxicity assessment of Irish marine sediments: validation of established marine bioassays for the monitoring of Irish marine sediments. Environ Int 34(7):1023–1032 Macleod CK, Crawford CM, Moltschaniwskyj NA (2004) Assessment of long term change in sediment condition after organic enrichment: defining recovery. Mar Pollut Bull 49:79–88 Mantzavrakos E, Kornaros M, Lyberatos G, Kaspiris P (2007) Impacts of a marine fish farm in Argolikos Gulf (Greece) on the water column and the sediment. Desalination 210:110–124 Marıń A, Montoya S, Vita R, Marıń-Guirao L, Lloret J, Aguado F (2007) Utility of sea urchin embryo-larval bioassays for assessing the environmental impact of marine fish cage farming. Aquaculture 271:286–297

Integrated ecotoxicological assessment of marine sediments Martıń-Dıáz ML, Blasco J, Sales D, DelValls A (2008) Field validation of a battery of biomarkers to assess sediment quality in Spanish ports. Environ Pollut 20:1–10 McPherson C, Chapman PM, deBruyn AMH, Cooper L (2008) The importance of benthos in weight of evidence sediment assessments—a case study. Sci Total Environ 394(2–3):252–264 Mendiguchıá C, Moreno C, Mańuel-Vez MP, Garcıá-Vargas M (2006) Preliminary investigation on the enrichment of heavy metals in marine sediments originated from intensive aquaculture effluents. Aquaculture 254:317–325 Montero N, Belzunce-Segarra MJ, Del Campo A, Garmendia JM, Ferrer L, Larreta J (2011) Integrative environmental assessment of the impact of Pasaia harbour activities on the Oiartzun estuary (southeastern Bay of Biscay). J Mar Syst (in press). doi: 10.1016/j.jmarsys.2011.06.002 Morales-Caselles C, Kalman J, Riba I, DelValls TA (2007) Comparing sediment quality in Spanish littoral areas affected by acute (Prestige, 2002) and chronic (Bay Of Algeciras) oil spills. Environ Pollut 146:233–240 Muxika I, Borja A, Bald J (2007) Using historical data, expert judgement and multivariate analysis in assessing reference conditions and benthic ecological status, according to the European Water Framework Directive. Mar Poll Bull 55:16–29 Ocanã-Martıń A, Sanchez-Tocino LL, Lo´pez-Gonza´lez S, Viciana JF (2000) Guıá Submarina de invertebrados no artro´podos, 2nd edn. Granada, Comares, p 512 Papageorgiou N, Sigala K, Karakassis I (2009) Changes of macrofaunal functional composition at sedimentary habitats in the vicinity of fish farms. Estuar Coast Shelf Sci 83:561–568 Pearson TH, Black KD (2001) The environmental impacts of marine fish cage culture. In: Black KD (ed) Environmental impacts of aquaculture. Academic, Sheffield, pp 1–27 Pearson TH, Rosemberg R (1978) Macrobenthic succession in relation to organic enrichment and pollution of the marine environment. Oceanogr Mar Biol 16:229–311 Pe´rez E, Blasco J, Sole´ M (2004) Biomarker responses to pollution in two invertebrate species: Scrobicularia plana and Nereis diversicolor from the Cadiz bay. Mar Environ Res 58:275–279 Peters C, Becker S, Noack U, Pfitzner S, Bu¨low W, Barz K, Ahlf W, Berghahn R (2002) A marine bioassay test set to assess marine water and sediment quality-its need, the approach and first results. Ecotoxicology 11:379–383 Pusceddu A, Fraschetti S, Mirto S, Holmer M, Danovaro R (2007) Effects of intensive mariculture on sediment biochemistry. Ecol Appl 17:1366–1378 Ramos-Go´mez J, Martıń-Dıáz ML, DelValls TA (2009) Acute toxicity measured in the amphipod Ampelisca brevicornis after exposure to contaminated sediments from Spanish littoral. Ecotoxicology 18:1068–1076 Read P, Fernandes T (2003) Management of environmental impacts of marine aquaculture in Europe. Aquaculture 226:139–163 Riba I, DelValls TA, Forja JM, Go´mez-Parra A (2003) Comparative toxicity of contaminated sediment from a mining spill using two amphipods species: Corophium volutator (Pallas, 1776) and Ampelisca brevicornis (A. Costa, 1853). Bull Environ Contam Toxicol 71:1061–1068

1011 Riba I, Casado-Martıńez C, Forja JM, DelValls TA (2004) Sediment quality in the Atlantic coast of Spain. Environ Toxicol Chem 23(2):271–282 Riedl R (1986) Fauna y flora del mar Mediterrańeo. Omega, Barcelona, p 858 Rodrigues AM, Meireles S, Pereira T, Gama A, Quintino V (2006) Spatial patterns of benthic macroinvertebrates in intertidal areas of a Southern European estuary: the Tagus, Portugal. Hydrobiologia 555:99–113 Sapkota A, Sapkota AR, Kucharski M, Burke J, McKenzie S, Walker P, Lawrence R (2008) Aquaculture practices and potential human health risks: current knowledge and future priorities. Environ Int 34:1215–1226 SEPA (2005) Scottish Environment Protection Agency. Regulation and monitoring of marine cage fish farming in Scotland. A manual of Procedures. Standard Monitoring Survey. Version 1.6. http://www.sepa.org.uk/guidance/fishfarmmanual/manual.asp Shannon CE, Weaver W (1949) The mathematical theory of communication. The University of Illinois Press, Urbana Solan M, Cardinale BJ, Dowing AL, Engelhardt KAM, Ruesink JL, Srivastava DS (2004) Extinction and ecosystem function in the marine benthos. Science 306:1177–1180 Sole´ M, Kopecka-Pilarczyk J, Blasco J (2009) Pollution biomarkers in two estuarine invertebrates, Nereis diversicolor and Scrobicularia plana, from a Marsh ecosystem in SW Spain. Environ Int 35:523–531 Tello A, Corner RA, Telfer TC (2010) How do land-based salmonid farms affect stream ecology? Environ Pollut 158:1147–1158 Thorson G (1957) Bottom communities (sublittoral or shallow shelf). In: de Hedgpeth JH (ed) Treatise on marine ecology and paleoecology, vol 1. Ecology Mem Geol Soc Am 67:461–534 Tomassetti P, Porrello S (2005) Polychaetes as indicators of marine fish farm organic enrichment. Aquac Int 13:109–128 Tomassetti P, Persia E, Mercatali I, Vani D, Marussso V, Porrello S (2009) Effects of mariculture on macrobenthic assemblages in a western mediterranean site. Mar Pollut Bull 58:533–541 Tovar A, Moreno C, Mańuel-vez MP, Garcıá-Vargas M (2000a) Environmental impacts of intensive aquaculture in marine waters. Water Res 34:334–342 Tovar A, Moreno C, Mańuel-vez MP, Garcıá-Vargas M (2000b) Environmental implications of intensive marine aquaculture in earthen ponds. Mar Pollut Bull 40:981–988 US, EPA (United States Environmental Protection Agency) (1994) Methods for assessing the toxicity of sediment-associated contaminants with estuarine and marine amphipods (EPA/600/ R-94/025). Office of Water, Washington Usero J, Morillo J, El Bakouri H (2008) A general integrated ecotoxicological method for marine sediment quality assessment: application to sediments from littoral ecosystems on Southern Spain’s Atlantic coast. Mar Pollut Bull 56:2027–2036 Vezzulli L, Moreno M, Marin V, Pezzati E, Bartoli M, Fabiano M (2008) Organic waste impact of captured-based Atlantic bluefin tuna aquaculture at an exposed site in the Mediterranean Sea. Estuar Coast Shelf Sci 78:369–384

123