Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
ICES CM 2010/H:01 Note be cited without prior reference to the author Benthic Indicators for assessment the quality status of coastal soft-bottom communities: response of different indices to diverse human pressures in the North-eastern Atlantic Ocean and Mediterranean Sea Jean-Claude Dauvin1, Sandrine Alizier1, Céline Rolet1, Ali Bakalem2, Gérard Bellan3, Jose Luis Gomez Gesteira4, Samir Grimes5, de-la-Ossa-Carretero6 & Yoana Del-Pilar-Ruso6 1 Université de Lille 1, CNRS UMR 8187 LOG, Station Marine de Wimereux, 28 Avenue Maréchal Foch, B.P. 80, F-62930 Wimereux, France. Phone + 33 3 21 99 29 20; Fax + 33 3 21 99 29 01; e-mail: [email protected], [email protected], [email protected] 2 Ecole Nationale Supérieure Agronomique (ENSA), Avenue Hassan Badi, 16200, El Harrach, Alger, Algérie ; e-mail : [email protected] 3 Centre d’Océanologie de Marseille, UMR CNRS DIMAR 6540, Université de la Méditerranée, Station marine d’Endoume, Rue Batterie des Lions, F-13007 Marseille; e-mail: [email protected] 4 Centro Tecnológico del Mar – Fundación CETMAR, Eduardo Cabello, s/n, 36208 Vigo. Spain; e-mail: [email protected]. 5 Ecole Nationale Supérieure des Sciences de la Mer et de l’Aménagement du Littoral (ESSMAL), BP 19, Campus universitaire de Dely Brahim, Bois des Cars, Alger, Algérie ; email [email protected] 6 Department of Marine Sciences and Applied Biology, University of Alicante, Ap 99 E03080 Alicante. Spain; e-mail: [email protected], [email protected] Abstract. The interest on benthic indicators for soft-bottom marine communities has dramatically increased after a rather long term of relative stagnation due to the need of new tools to assess the status of marine waters within Clean Act and the Framework Water Directive. Our expertise on benthic communities had permitted to accumulate a large amount of data from diverse unpolluted and polluted conditions (accidental oil spill, sewage, longterm anthropogenic estuarine constraints) in tidal estuaries, harbours, and coastal shelf from Atlantic Ocean and Mediterranean Sea, i.e. English Channel, Atlantic and Mediterranean coast of Spain, Marseilles Gulf and Algerian coats. We tested several kinds of benthic indicators on diverse available datasets, such as “sentinel species” which is a particular species which by its presence or its relative abundance would have for the observer a role of “warning” vis–à-vis possible unbalances in the surrounding environment or distortions in the functioning of the community. We have also checked several indices: Shannon-Wiener H’ diversity, AMBI and BO2A indices according to the classification of species into Ecological Groups and ITI according to the classification of species into trophic groups. We discuss on the selection and use of Biological Indicators as “objective” or “subjective” alternatives to assess diagnostic for soft-bottom communities and to characterize the state of an ecosystem 2 and to bring to the fore, as early as possible, its modifications, natural or provoked. We recommend to be staying pragmatic and act to transfer simple method for the research consultancies which are in charge of the assessment of the benthic quality in numerous impact studies. Keywords: Benthic indicators; soft-bottom communities; subjectivity; objectivity; human pressures; comparison of sites Contact author: Jean-Claude Dauvin, Université de Lille1, Station Marine de Wimereux, UMR LOG, 28 avenue Foch, BP 80, F-62930 Wimereux, France. Phone: 33 3 21 99 29 20; email: [email protected] Introduction Since the publication of the Water Framework Directive in 2000 (WFD, 2000/60/EC), the research of more efficient indicators to qualify the water quality of the water body masses had generated a dramatically increase of the numbers of papers. Moreover several workshops, conferences and colloquiums had been specifically organised on this topic or special sessions in other international forums participated to this debate and they were the sources of a profusion of more or less new tools (see Devlin et al., 2007; ICES, 2008; Pinto et al., 2009; Borja et al., in press). Among bio-indicators, five biological compartments were retained in the WFD, i.e. phytoplankton, macroalgae, angiosperms, macrozoobenthos and fish. Among them, the development of macrobenthic indicators was among the most significant. This was probably due to the fact that this component present a lot of advantages to assess ecological quality: 1) they are relatively sedentary; 2) they have relatively long life-spans; 3) they comprise diverse species that exhibit different sensitivities or tolerances to stress; and 4) they play an important role in cycling nutrients and materials between the underlying sediments and the overlying water column (Dauvin, 2007; Bellan, 2008; Borja et al., 2008). Nevertheless, Rakocinski and Zapfe (2005) have underlined several disadvantages of the existing benthic indices based on benthic organisms: 1) they represent a static expression of an ecological condition, 2) they are not explicitly linked to changes in ecological function, 3) they may not be specific with respect to different kinds of stressors, 4) they are subject to underlying taxonomic changes across estuarine gradients, 5) their use can be labour intensive, and 6) they are not applied consistently across bio-geographic provinces. This last point was recently underlined by Grémare et al. (2009) when these authors compared the efficiency of two indices, AMBI and BQI, at a European scale. However, the use of macrobenthos as biological indicators had begun several years ago before the WFD. One of the foundation 3 studies was those of Reish (1959) on the Los Angeles and Long Beach harbours five decades ago. The concept of sensitivity or resistance of macrobenthic species had greatly inspired by the paradigm of macrobenthos succession with respect to organic enrichment and pollution of the marine environment (Pearson and Rosenberg, 1978). This paper was the origin of numerous discussions including the conception of indicators and indices used in soft-bottom macrobenthic communities, as well as these developed within the WFD. But the increase of organic matter was not the unique source of pollution in the sediment, i.e. other sources such as metals, hydrocarbons, deposit of dredging…, can also affect the structure and the functioning of benthic communities. This was the limit of the Pearson and Rosenberg concept, and of the following developed according to the response of benthic species to the increasing of organic matter, i.e. most of the benthic indicators. These indicators were based on the species classification in several ecological groups from the more sensitive to the more tolerant species versus the increase of organic matter. Moreover classifications of species in ecological groups have been based in part on subjective or objective expertise’s; so most of the biological indicators and consequently pollution indices created are doubtful. But, this reality do not dissuade the development, adaptation and creation of numerous new indices notably since the beginning of the 2000’s in spite of several advertisement on their tautology, marked by Diaz et al. (2004). Nowadays, some papers promote to simplify the approach, and envisage taking into account the taxonomy sufficiency (Dauvin et al., 2009; Dauvin and Ruellet, 2009) or the expert judgement (Weisberg et al., 2008; Teixeira et al., 2010). The possibility to use species classifications, in the concept of the community or biocenoses, and in the primary pollution studies on the benthic communities, was also an alternative to give a diagnostic to the ecological quality status of benthos. Our expertise on benthic communities had permitted to accumulate a large amount of data from diverse unpolluted and polluted conditions (accidental oil spill, sewage, long-term anthropogenic estuarine and their local or regional adaptation including the need to take into account the special environment which occurs in the transitional water bodies) in various environments, from tidal estuaries, harbours, and coastal shelf from Atlantic Ocean and Mediterranean Sea, i.e. English Channel, Atlantic and Mediterranean coast of Spain, Marseilles Gulf and Algerian coats. The objectives of this paper, after a selection of a set of data that permit to visualize several shallow and harbours unpolluted or polluted situations, are: 1) to test several kinds of benthic indicators, i.e. “sentinel species”, Shannon H’ diversity, AMBI, BO2A and ITI, on selected 4 available datasets; 2) to test if the best professional judgment (BPJ) agree on condition assessment between regions and natural and anthropogenic constraints, and 3) to discuss on the subjectivity and objectivity involved in using benthic indicators. Materials and methods Sampling sites and selection of data (see figure 1 for the locations of the sampling sites) Figure 1. Locations of the sampling sites in the North-eastern Atlantic Ocean and western part of the Mediterranean Sea; BS, Bay of Seine; BM, Bay of Morlaix; GA, Galicia, CA, Castellon coast; MA, Marseilles harbour; CO, Cortiou emissary; GH Ghazaouet; OR, Oran; AR, Arzew; BI, Bou Ismail; AL, Algiers; BE, Bejaia JI; Jijel; FE, Fetzara; AN, Annaba. Bay of Seine The Bay of Seine and the Seine estuary, which is the largest megatidal estuary in the English Channel, covering approximately 150 km2 at high tide, is one of the more human-impact estuaries of the Northeastern Atlantic Ocean. This is due to the accumulation of disturbances within dykes and embankments to ameliorate the navigation. It reduces considerably the surface of the intertidal zone, and the presence of the more inhabited watershed in France which provokes a lot of pollution with high levels of nutrients, particularly nitrate, heavy metals, mainly cadmium and lead, PAHs, PCBs and pesticides. This cocktail helps to place the Seine estuary among the most contaminated of estuaries in the Europe (Dauvin et al., 2007). Two main subtidal communities have been identified in the lower part of the Seine 5 estuary: 1). An Abra alba-Pectinaria koreni muddy sand community occupies the external part of the estuary and the entrance to the North and South Channels, and shows high diversity, abundances and biomasses and 2). A Macoma balthica community inhabits inner subtidal bottoms in the Channel located in the polyhaline and mesohaline zones of the estuary with low species diversity, abundances and biomasses. During the COLMATAGE 2008 campaigns in September, 11 sites (5-20 m) covering all the salinity gradient of the estuary and a northern-southern transect were selected for this study. Five replicates with a Van Veen grab were sampled at each site; the sediment was sieved on 1 mm mesh. Bay of Morlaix In 1977, a benthic survey of temporal changes was undertaken to perform a long-term monitoring of two soft-bottom communities from the Bay of Morlaix (Pierre Noire, PN and Riviere de Morlaix, RM; Western English Channel, France). The PN site is located in the eastern part of the Bay of Morlaix, Brittany (48°42, 50'N; 3°51, 96'W), 17 m depth at low tide on a Abra alba-Aponuphis bilineata fine sand community. The RM site (48°39, 27'N; 3°52, 08'W), 10 m depth at low tide, is on a muddy sand Abra alba-Melinna palmata community in an area under freshwater influence. Both sites were polluted by hydrocarbons from the Amoco Cadiz wreck in April 1978. But, the spill's impact was more pronounced at Pierre Noire where amphipods of the genus Ampelisca had dominated prior to the spill (Dauvin, 1998). These amphipods disappeared in April 1978 just after the sediment was contaminated by hydrocarbons, and the recolonisation of the destroyed population took 10 years (Dauvin, 1998). Though a small number of amphipods did disappear from the Rivière de Morlaix site, which before the spill had been dominated by polychaetes, particularly by Melinna palmata and Chaetozone gibber in such estuarine environment with high level of organic matter in natural condition (Dauvin, 2007), the impact of the Amoco Cadiz oil spill was insignificant (Dauvin, 2000). Ten replicates with a Smith McIntyre grab were sampled at each site; the sediment was sieved on 1 mm mesh. For each site, the winter (February-March) and summer (August) sampling from August 1977 to March 1983 (12 dates) were selected for a total of 24 observations from the Bay of Morlaix. Galicia, Ares-Betanzos Ria (Northwest Spain) The Ares-Betanzos Ria from the Northwest Spain, in Galicia, is a double estuarine system covering a surface of 73 km2. Depths range from 2 to 43 m. Sediment is distributed in three 6 principal zones. The inner zone with a depth of < 10 m is dominated by muddy sediments with high silt and clay content; the central zone, whose depth ranges between 10 m and 30 m, is dominated by medium to fine sands, and the mouth, the deepest and most exposed zone of the ria, is dominated by gravel and coarse sand with a notable shell input. Sampling sites were initiated three days after that the Greek tanker Aegean Sea scratched on 3 December 1992 and released 79,000 tons of “brent blend” crude oil in an accident occurring just outside the A Coruña harbour. Two sites (10-15 m in depth) X (43°23.90’N; 8°14.15’W) and Z (43°23.85’N; 8°15.75’W) were selected for this study ; they are located in the sandy central zone of the ria, in an ecotone between the Abra alba and Venus gallina communities (Gomez Gesteira and Dauvin, 2005). Reported responses of infauna after an oil spill include very high initial mortalities in species sensitive to hydrocarbons, such as crustaceans and especially amphipods, and their subsequent disappearance. This process, comparable to those observed in the Amoco Cadiz oil spill in Brittany, is followed, 1-3 years after the spill, by a significant increase in the abundance of opportunistic species, such as polychaetes, which proliferate due to increased organic matter (Gomez Gesteira and Dauvin, 2005). At each date, six samples were taken with a 1/127 m2 box corer; the sediment was sieved on 1 mm mesh. The first sampling in December 2002 and the last sampling in November 2006, and winter (MarchApril) and summer (July-August) were selected for a total of ten dates at each site and a total of 20 observations. Castellon coast (Mediterranean north-eastern Spain) This study area is located on the Castellon coast (north-eastern Spain); five locations affected by sewage outfalls along 40 km of coast were analysed. These outfalls correspond to the villages of Vinaroz (VIN), Benicarló (BEN), Peñíscola (BEN), Alcossebre (ALC) and Torreblanca (TOR). The mean length of the pipelines is 2,138 m discharging at mean depth of 14.8 m in fine-sand communities, with homogeneous bottom sediment and uniform benthic communities. The associated macrofauna were characterized by high abundances of the bivalve Spisula subtruncata and the tanaid Apseudes latreilli (de-la-Ossa-Carretero et al., 2008, 2009, 2010). For each location, three distances from the discharge (0, 200 and 1,000 m) were sampled in summer, coinciding with the highest rate of sewage disposal, during July of 2006. Three Van Veen grab samples (400 cm2) were collected at each site, for a total of 15 situations and the sediment was sieved through a 0.5 mm screen. 7 Marseilles area Marseilles harbour A total of 12 situations in the Marseilles harbour was selected according to a pollutedunpolluted gradient in summer 1965 and 1966, i.e. three sites in the polluted area (inner basins), three in a moderated polluted area (inner navigation channel), the other in the larger basins of the northern harbour, three in a slightly polluted area at a depth lower than 15 m and finally three in a slightly polluted area at a depth higher than 15 m (Bellan, 1967a). Samples were obtained with a size one orange peel bucket grab which collects approximately 5 dm3 of sediment over a 0.12 dm2 surface. At this period the harbour, and more specially, the inner basins was severely polluted by hydrocarbons, plants, and few small rivers rather “open- air sewer”. These conditions are now no longer bearable. The benthic assemblages in the harbours were very similar with the Cortiou area ones, function of depth and nature of the bottom. It is particularly interesting to note that these assemblages were similar, if not identical, to those observed in the Algerian harbours. Cortiou A total of 12 situations was selected for the Cortiou emissary, six near the emissary, i.e. a station sampled in 1965, then the I7 site sampled regularly during the summer in 1989, 1994, 1995, 1997 and 2001, and six at a certain distance of the emissary, i.e. three sites F2, E6 and D2 sampled respectively in 1976, 1980, 1987, 1994. Sampling was made according the same method than in the Marseilles harbour. The selected samplings have to be referred in consideration of particular modifications with the modifications of the volume and nature of the discharges (Bellan, 1967 b). From the 1965-1966 samplings, the spatio-temporal changes showed that the area of the polluted zone area increased by a factor of 3 and the subnormal (transitional) zone by a factor of 10 between 1970 and 1980. After the construction of the primary treatment plant in 1987, the macrobenthic populations of the polluted zone, near the emissary, decreased sharply in terms of the number of individuals. The composition of the assemblages began to change in 1992-1993 indicating a progressive recovery of areas previously degraded from the polluted zone, to a community resembling the subnormal (transitional) zone. These changes in the benthic populations indicates that construction of the treatment plant has greatly reduced the contamination in the area and led to the drastic 8 reduction of the polluted zone and the recovery of the benthos around the Cortiou region, specially in the sites F2, E6 and D2. (Bellan et al., 1999). The recent (2006), servicing of a secondary-treatment plant seems to have increased these patterns. Algerian coast Fine sand communities Benthic samplings were made during the 1980s and the beginning of the 1990s in the shallow (9 and 40 m of depth) fine sand communities, with a sediment median between 85 and 175 µm containing a low percentage of fine particles, mainly lower than 5 %, in seven bays along the Algerian coast (Bakalem, 2008). The sandy species characterizing this community were the polychaetes Scoletoma impatiens and Owenia fusiformis, the phoronid Phoronis psammophila, the bivalves Spisula subtruncata and Venus gallina, and the amphipods Ampelisca brevicornis and Urothoe spp. Though the fine sand communities occupy the shallow depths in all the Algerian bays, their surfaces vary strongly from one bay to another: they occupy only a small area in the Bay of Oran, are well developed in Jijel and Algiers Bays, and occupy a small strip parallel to the coast in the other bays (Bakalem et al., 2009). A 0.1 m-2 Van Veen grab was used in Algiers and Jijel Bays, while a 0.1 m-2 Smith McIntyre grab was used in the five other bays. All the sieving was done with a 1 mm mesh. A total of 16 samplings was selected according to the size of the bays and representative to the entire Algerian coast, i.e. two samplings in the Fetzara, Jigel, Bejaia, Bou Ismail, Arzew and Oran bays and four sampling in the Algiers Bay. Harbours Twelve harbours were prospected during the 1995-2001 period; all the samples (3-35 m) were taken in the summer period between 15 June and 15 August. The geographical limits of the sampling were 35°6'N-2°9'00”W near the Moroccan border and 36°54'N-8°26'30”E near the Tunisian border (Grimes et al., in press, in preparation). At each site, two replicates of a Van Veen grab covering a total substratum surface of 0.25 m2 were taken; the sediment was sieve on a 1 mm mesh. A total of 14 samplings was selected in this study, four in the Annaba and Algiers harbour, and two in the Bejaia, Ghazaouet and Oran harbours covering the entire Algerian coast (Grimes et al., in press). 9 The studies conducted by the METAP (1993, 1994), showed that heavy metal pollution affected differently the various harbours explored. Algiers, Oran and Ghazaouet harbours were the most affected. The first two were polluted by high loads inside and outside the harbours while in Ghazaouet harbour the concentrations of Zn, Pb, Hg and Cd were more important inside harbour sediments. In contrast, the port of Bejaia, Cd concentrations were very important in surface indicating recent inputs of Cd. The ISMAL studies, according to a synthesis made by Grimes (2003) verified the METAP (1994) results and showed that in the majority of the cases high concentrations of metal contaminants were revealed in surface sediments, some exceeding accepted standards and proved that sources of pollution were very localized. High Zinc concentration found in the Ghazaouet harbour was in relation with electrolyse of zinc factory. High concentrations were measured in the sediments of Oran Harbour, particularly for mercury: 0.47-1.21 μg/g, copper: 64.95-119.7 µg/g, and zinc 161.75262.56 µg/g (217.85 ±64.95-119.7 µg/g). The Algiers harbour showed critical levels, especially for copper: 112.01± 39.93 µg/g, zinc: 338.342±103.737 µg/g and cadmium: 0.883±0.225 µg/g. These concentrations ranked the port of Algiers as a risk zone. The maritime traffic, port activity, and urban discharges from the city of Bejaia flowing inside the harbour and biogenic particles were main sources of inorganic pollution. Zinc contamination was detected (186.93±61.11 µg/g). In Bejaia harbour, contamination reflected the combined influence of uncontrolled sewage and industrial discharges (ENCG UP7, ECOTEX, UNGG UP8) in Oued Soummam and the harbour towards the marine environment. Metal pollution in this area was mainly expressed in the copper and manganese. The organic material from sewage of the city of Bejaia affected the quality waters of the Gulf of Bejaia. The region of Annaba is polluted by heavy metals as well as hydrocarbons. Sources of water pollution came mainly from industrial discharges in the harbour area and the urban sewage in the major rivers. All the factory releases were discharged to the Gulf of Annaba, through both Oueds Seybouse and Boudjemâa. Assessment the quality status Benthic indicators The benthic indicators used in this paper can be divided into four categories: - Indicators strictly based on ecological groups. The AMBI (AZTI Marine Biotic Index) developed by Borja et al. (2000), which analyzes the proportions of five ecological groups 10 (using the species list published by the AZTI web site in February 2010) according to a gradient of organic matter enrichment (Pearson and Rosenberg, 1978) was calculated following the guidelines of Borja and Muxika (2005). The BO2A (Benthic Opportunistic Annelida Amphipods index) (Dauvin and Ruellet, 2009) respecting the principle of taxonomic sufficiency using only two well-known zoological groups as indicator species, consequently limiting the misclassification of taxa caused by too many ecological groups was also calculated. BO2A was an adaptation of the BOPA (Benthic Opportunistic Polychaetes Amphipods index) developed by Dauvin and Ruellet (2007) for the transitional water body masses. - Indicators strictly based on trophic groups. The primary indicator in this group is the ITI (Infaunal Trophic Index; Mearns and Word, 1982), which holds that the ecological quality of the community increases with the dominance of suspension feeders and decreases with the dominance of subsurface deposit feeders. - Indicators strictly based on diversity. The main indicator in this category is H’ (the Shannon Index with log2) proposed by Shannon (1948): its use was based on the theory of information, as proposed in the original text. The five EcoQS (ECOlogical Quality Status) classes suggested by the WFD were also used in this study: high for unpolluted sites, good for slightly polluted sites, moderate for moderately polluted sites, poor for heavily polluted sites, and bad for extremely polluted or azoic sites. The thresholds used in this note are reported in the Table 1 (Ruellet and Dauvin, 2007), i.e. Borja et al. (2000) for AMBI, Vincent et al. (2002) for H’ indicator and those defined by dela-Ossa-Carretero and Dauvin (submitted) for the BO2A. For ITI, only three EcoQs were identified: degraded for sites with a dominance of subsurface deposit feeders; perturbed for sites showing miscellaneous trophic groups, such as subsurface deposit feeders, deposit feeders and suspensivores; and normal for sites with a dominance of suspensivores. Table 1. Threshold values between the five ecological status considered for the selected benthic indicators (see text for definitions and references for the levels of the thresholds) of AMBI, BO2A, H’, and ITI High Good Moderate Poor Bad AMBI BO2A H’ ITI 0-1.2 0.00000-0.01950 +∞-4.00 1.2-3.3 0.001951-0.13100 4.00-3.00 Normal 60-100 3.3-4.3 0.13101-0.19804 3.00-2.00 Perturbed 30-60 4.3-5.5 0.19804-0.26155 2.00-1.00 Degraded 0-30 5.5-7.0 0.26155-0.30103 1.00-0.00 11 Sentinel species and other categories of species Dauvin et al. (2010) have recently reviewed the definition of different terms used to characterize the species in the field of benthic indicators, i.e. sensitive species, tolerant species, opportunistic species, indifferent species, indicative species, indicator species, pollution indicator species, and finally sentinel species which is a particular species which by its presence or its relative abundance "warns" an observer about possible imbalances in the surrounding environment and/or alterations of the community functions. We have chosen to take into account in this study two kinds of sentinel species, those which indicated an increase of organic matter such as Capitella capitata complex, Malacoceros (Scolelepis) fuliginosa, Neanthes (Nereis) caudata complex, Cirriformia (Audouinia) tentaculata, Schistomeringos (Dorvillea, Staurocephalus) rudolphi among the polychaetes and Corbula gibba among the bivalves, and the amphipods Ampelisca species which are known to be very sensitive to disturbance. Best Professional Judgement (BPJ) Recent assessments of benthic macrofauna community conditions using "Best professional judgement" (BPJ) (Weisberg et al., 2008; Teixeira et al., 2010) has revealed that experts, based on their experience, are able to evaluate the quality status of benthic samples into four categories with a high degree of agreement and to rank samples from best to worst condition (Teixeira et al., 2000). There is also good agreement on the condition category, though the agreement was better for samples at the extremes of the disturbance gradient. The experience in benthic monitoring of four experts (AB, GB, JCD, and SG) who exercise ranged from 15 to 50 years have assigned blindly each of the 124 sites to one of four conditions categories on narrative description as suggested in Teixeira et al. (2010), 1) unaffected community (high and good states of the WFD), 2) marginal deviation from unaffected (moderate status of the WFD), 3) affected (poor status of the WFD), and 4) very affected (bad of the WFD) (only the lists of species with the number of individuals of each species were given without other information’s of the sample locations). The four experts were also asked to identify the criteria used in their judgement as: 1) very important, 2) important but secondary, 3) marginally important and 4) useful, but only to interpret other criteria. Seven criteria used by experts to categorize samples are: 1) presence of opportunistic species; 2) presence of tolerant species; 3) presence of sensitive species; 4) number of taxa 12 present in the sample; 5) total abundance in the sample; 6) abundance dominance patterns by opportunistic species and 7) abundance dominance pattern by sensitive species. Statistical analysis A Mann–Whitney–Wilcoxon (MWW) test which is a non-parametric test was used for assessing whether two series of diagnostics have equally values, i.e. there was no significant diagnostic (EcoQS) between the assessments (Scherrer, 1984). MWM tests were made between the assessment of the benthic indicators (H’ versus AMBI …) for paired samples and between the diagnostic of the four experts (p level <0.1%) for independent samples. For the BPJ a weighted Kappa analysis (Cohen, 1960; Landis and Kosch, 1977) was used to analyse the agreement among all the experts and mean judgement condition with respect to each expert. The methodology proposed by Borja et al. (2007) was employed. The equivalence table from Monserud and Leemans (1992) was used to establish the level of agreement. In addition, since the importance of misclassification is not the same between close categories (e.g., unaffected and affected) as between distant categories (e.g., between unaffected and severely affected), we chose to apply Fleiss–Cohen weights (Fleiss and Cohen, 1973) to the analysis to decrease importance of misclassification between close categories and increase importance between distant categories. Results General pattern of the dataset Due to the very different origins of the records, from severe perturbed environments to normal conditions, but also to different types of sampling gears and surface sampled in each site, the specific richness (SR) and number of collected specimens varies strongly between the data, i.e. 124 selected samples (Annexe; Figure 2). In fact, the SR varies from 0 in the navigational channel of the Seine estuary in an azoic zone due to the continuous dredging to access to the Rouen harbour which is located at 120 km off the sea, to 133 in August 1980 and August 1982 at the Pierre Noire site (Bay of Morlaix). SR showed a Gaussian form and most of the samples showed an intermediate SR included between 30 and 60 (Figure 2). The number of individuals also varies from 0, in the same site that SR in the Seine estuary, to 37,658 in August 1977 at the Pierre Noire site. The samples showed two maximums according their 13 abundances, i.e. about 30 % of the samples with a number of individuals included between 100 and 500 and about 30 % of the samples with a number of individuals included between 1,000 and 5,000. As suggested before by Dauvin and Ruellet (2007), the sites with a number of individuals > 20 were excluded in the assessment, i.e. three sites, two from the navigational channel of the Seine estuary and one in the Marseilles harbour where Capitella capitata was the single sampled species with six specimens. So, the total numbers of considered sites in our statistical analyses were 121 on a total of 124. Figure 2. Repartition of the 124 selected samples, a) according to the species richness and b) according to the number of individuals. Assessment using benthic indicators The figure 3 showed the number of samples corresponding to the five EcoQS given by H’, AMBI and BO2A. They were classified mainly as High and Good status (more than > 50 %), Poor and Bad Status were less represented; they were mostly located in the Marseilles and Algerian harbours and the Cortiou emissary (Annexe; Table 2). H’ appeared the more optimist index, following by AMBI and BO2A which gave more Poor and Bad status than 14 AMBI. For the ITI, about 75% of the samples showed an intermediate perturbed status, while the assessment identified a very low number of degraded samples (Figure 4). Figure 3. Number of samples corresponding to the five EcoQS given by H’, AMBI and BO2A (total of 121 samples). Figure 4. Number of samples corresponding to the three EcoQS given by ITI (total of 121 samples). In a first step, if the five EcoQS were taken into consideration, the MWW tests for paired samples showed that there were no significant differences between the assessment given by H’, AMBI and BO2A, therefore there was a good concordance among the three indices. In a second step, to compare with ITI, the High and Good status on one part (corresponding to the Normal status of ITI), and the Poor and Bad status on the other part (corresponding to the Degraded status of ITI) were pooled, while the Moderate status was not changed and compared with the Perturbed status of ITI. The MWW tests for paired samples showed that 15 only H’ and BO2A gave similar status (i.e. no significant difference between their assessment) while all comparisons gave significant different assessments. Table 2 illustrated clearly the divergence of diagnostic among the indices. Nevertheless, in the case of samples collected in some sites, such in Spain, the three indices (H’, AMBI an) BO2A gave very similar diagnostic, in spite of the Aegean oil spill in Galicia and in spite of the presence of sewage pollution along the Castellon coast. Diagnostics were more divergent for more stressed sites, i.e. in the Bay of Morlaix within the Amoco Cadiz oil spill, in the Marseilles and Algerian harbours, and in the area of the Marseilles Cortiou emissary. Table 2. Number of samples given by the different benthic indicators to assess the EcoQS in the different sites. HG, High and Good pooled assessment, M, Moderate assessment, BP, Bad and Poor pooled assessment, for Shannon H’, AMBI and BO2A. N, Normal, P, Perturbed, and D, Degraded for ITI. Site Bay of Seine, France N of samples 9 Bay of Morlaix, Pierre Noire, France 12 Bay of Morlaix, Rivière de Morlaix, France 12 Galicia, Spain 20 Castellon coast, Spain 15 Marseilles Harbour, France 11 Marseilles emissary, Cortiou, France 12 Algerian harbours, Algeria 14 Algerian coasts, Algeria 16 H’ 4 HG 3M 2 BP 9 HG 2M 1 BP 3 HG 7M 2 BP 20 HG 15 HG 2 HG 6M 3 BP 9 HG 1M 2 BP 5 HG 1M 8 BP 11 HG 3M 2 BP AMBI 9 HG 11 HG 1M 2 HG 10 M 20 HG 15 HG 7 HG 4 BP 10 HG 2 BP 4 HG 6M 4 BP 16 HG - BO2A 7 HG 2M 11 HG 1 BP 1HG 2M 9 BP 20 HG 15 HG 4 HG 4M 3 BP 7 HG 3M 2 BP 7 HG 2M 5 BP 16 HG - ITI 2N 6P 1D 3N 9P 12 P 1N 19 P 3N 12 P 2N 7P 2D 2N 9P 1D 3N 10 P 1D 5N 9P 2D 16 It is clear that the ITI based on the trophic pattern of the species with only three status was more divergent that other indices for the assessment of the EcoQS (Table 2); it was also linked with the fact that a large majority of perturbed (mean) status was given by this index. Sentinel species The percentages of two kinds of sentinel species were represented according to the EcoQS given by H’, AMBI, BO2A and ITI in the figures 5, 6 and 7 (only percentages > 1% were taken into account in the figures; see annexe for the raw data). Opportunistic species were plotted for the five or three EcoQS of the four indices. There was a general trend showing that low dominance was associated mainly with High and Good EcoQS while Poor and Bad EcoQS were associated with high dominance of sentinel opportunistic species indicating probably high level of organic matter in the sediment. This pattern was more accurate for AMBI than for H’ and BO2A. For ITI there was no apparent relationship. The samples showing high dominance of opportunistic species (i.e. > 50 %) can be classified as Normal or Perturbed, not systematically as Degraded. Conversely Ampelisca were present mainly in the case of High and Good EcoQS except two samples in the Bay of Morlaix which occurred after the Amoco Cadiz oil spill during postspill colonisation of the Ampelisca and a summer proliferation of the tolerant polychaete species Pseudopolydora pulchra when Poor or Bad EcoQS where recognized by H’ and BO2A. AMBI showed more accurate relationship than both indices H’ and BO2A. For ITI there was no apparent relationship, but the samples with high dominance of Ampelisca were Normal while Degraded status were never associated with the presence of Ampelisca. Best Professional Judgement (BPJ) Table 3 gave the four categories assigned by the four benthic experts to each of the 124 samples selected in the dataset. 1: unaffected (green); 2: marginal deviation from unaffected (yellow); 3) affected (orange) and 4 severely affected (red). The experts were in a perfect agreement for 51 samples (41%), and only one expert differed from the three others for 46 (37 %) other samples, so there was good agreement between experts in more than three cases on four. The result of the MWW tests for independent samples showed that they were no significant differences between the assessments given by the four experts: there was a high agreement among experts. Similarly, the results of Kappa agreement analysis (Table 4) 17 indicated a high agreement among experts, the experts had a “very good” agreement, and this agreement was “almost perfect” comparing each expert judgement with mean judgement condition. Shannon diversity index (H') a) High Good Moderate Poor Bad 0 20 40 60 80 100 Opportunistic species (%) b) AMBI Index High Good Moderate Poor Bad 0 20 40 60 80 100 Opportunistic species (%) c) BO2A Index High Good Moderate Poor Bad 0 20 40 60 80 100 Opportunistic species (%) Figure 5. Percentage of the sentinel opportunistic species (samples with a % > 1) in the samples versus the five Ecological Quality Status given by: a) Shannon H’; b) AMBI and c) BO2A. 18 Shannon diversity index (H') a) High Good Moderate Poor 0 20 40 60 80 100 Ampelisca species (%) AMBI Index b) High Good Moderate 0 20 40 60 80 100 Ampelisca species (%) c) BO2A Index High Good Moderate Poor Bad 0 20 40 60 80 100 Ampelisca species (%) Figure 6. Percentage of the sentinel sensitive Ampelisca species (samples with a % > 1) in the samples versus the five Ecological Quality Status given by: a) Shannon H’; b) AMBI and c) BO2A. 19 ITI Index a) Normal Perturbed Degraded 0 20 40 60 80 100 Opportunistic species (%) ITI Index b) Normal Perturbed 0 20 40 60 80 100 Ampelisca species (%) Figure 7. Percentage of the sentinel opportunistic species (samples with a % > 1) (a) and percentage of the sentinel sensitive Ampelisca species (samples with a % > 1) (b) in the samples versus the three Ecological Quality Status given by the ITI index. The table 5 gave the rank of the criteria used by the four experts. All the experts judged as very important the presence of opportunistic species, and only useful (4), the total abundance in the samples, and as very important or important the presence of sensitive species. Three of the experts judged the dominance by opportunistic species as very important, and the presence of tolerant species as important. For the other criteria, the experts had more divergent arrangement. 20 Table 4. Kappa values with level of agreement in parenthesis (lower left) for condition category assignments, and percentage of mismatch (upper right between expert and mean judgement) classification. Level of agreement: AP- “Almost perfect”, VG- “Very Good”, G“Good”. A1 A1 A2 A3 A4 Mean 0.80 (VG) 0.79 (VG) 0.82 (VG) 0.94 (AP) Percentage of mismatch A2 A3 A4 MEAN 34.68 34.68 27.42 14.52 39.52 30.65 30.65 0.77 (VG) 33.87 24.19 0.74 (VG) 0.78 (VG) 16.13 0.86 (VG) 0.86 (AP) 0.89 (AP) - Table 5. Criteria used by the four experts to rank and categorize species samples. 1) very important; 2) important but secondary; 3) marginally important and 4) useful, but only to interpret other criteria. Criteria A1 A2 A3 A4 Presence of opportunistic species 1 1 1 1 Presence of tolerant species 3 2 2 2 Presence of sensitive species 1 2 2 1 Number of taxa present in the sample 2 3 1 2 Total abundance in the sample 4 4 4 4 Abundance dominance patterns by opportunistic species 1 1 1 4 Abundance dominance pattern by sensitive species 1 2 2 3 21 Table 3. Condition categories assigned by the four experts (A1, A2, A3 and A4) to each of the 124 samples selected in the dataset. 1: unaffected (green); 2: marginal deviation from unaffected (yellow); 3) affected (orange) and 4 severely affected (red). BS: Bay of Seine; RM: Rivière de Morlaix, Bay of Morlaix; PN: Pierre Noire, Bay of Morlaix; X and Z, Galicia, 02, February…; /77, 1977…; Castellon coast, Vinaroz (VIN), Benicarló (BEN), Peñíscola (BEN), Alcossebre (ALC) and Torreblanca (TOR), V1, near the emissary, N1, 200 m from the emissary, N2, 1000 m from the emissary; C, Cortiou; II, II/III, III<15, III> 15, Marseilles harbour; ALG, Algiers; ARZ, Arzew; BEJ, Bejaia; FET, Fetzara; JIJ, Jijel; OR, Oran; HALG, Algiers harbour; ; HAN, Annaba harbour; HBEJ, Bejaia harbour; HGZ Ghazaouet harbour; HORA, Oran harbour. Samples BS2 BS7 BS10 BS11 BS15 BS24 BS28 BS40 BS55 BS56 BS58 RM08/77 RM02/78 RM08/78 RM02/78 RM08/79 RM02/80 RM08/80 RM02/81 RM08/81 RM03/82 RM08/82 RM03/83 PN08/77 PN03/78 PN08/78 PN03/79 PN08/79 PN03/80 PN08/80 PN03/81 A1 2 2 1 2 2 3 2 2 4 3 3 3 2 3 3 3 3 3 3 3 3 3 2 1 1 2 2 2 2 2 2 A2 1 2 1 1 1 3 2 1 4 1 1 3 2 3 2 2 2 3 3 3 2 2 2 1 1 2 2 2 2 2 2 A3 3 3 2 2 1 3 2 2 4 3 3 3 2 3 2 2 2 3 2 3 2 3 1 1 2 2 2 2 1 1 2 A4 2 2 1 1 1 3 2 2 4 3 3 2 2 3 2 2 2 3 3 3 2 2 2 1 1 2 1 2 2 2 2 Samples PN08/81 PN03/82 PN08/82 PN03/83 X12/92 X03/93 X08/93 X03/94 X09/94 X04/95 X08/95 X04/96 X07/96 X11/96 Z11/92 Z03/93 Z08/93 Z03/94 Z09/94 Z04/95 Z08/95 Z04/96 Z07/06 Z11/06 VIN.V1 VIN.N1 VIN.N2 BEN.V1 BEN.N1 BEN.N2 PEN.V1 A1 2 2 2 1 1 2 1 2 1 1 1 1 1 1 1 2 1 2 2 1 1 1 1 1 2 2 1 2 1 1 2 A2 1 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 A3 1 1 2 1 1 3 1 2 1 1 1 1 1 1 1 2 2 1 2 1 1 1 1 1 2 1 1 1 1 2 2 A4 1 1 2 1 1 2 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 2 1 1 3 1 1 2 Samples PEN.N1 PEN.N2 ALC.V1 ALC.N1 ALC.N2 TOR.V1 TOR.N1 TOR.N2 C65C6 C89I7e C94I7e C95I7e C96I7e C01I7e C76F2 C76E6 C76D2 C94F2 C94E6 C94D2 II II II II/III II/III II/III III<15 III<15 III<15 III>15 III>15 A1 1 2 1 2 1 1 1 2 4 4 2 3 3 4 3 2 1 1 1 2 4 4 4 4 3 3 3 3 4 2 3 A2 1 1 1 1 1 1 1 1 4 4 1 3 1 1 2 1 1 1 1 1 4 4 4 4 4 4 3 2 3 1 1 A3 2 2 1 2 1 1 2 2 4 4 2 2 3 2 3 1 3 2 1 2 4 4 4 4 4 3 3 2 4 3 3 A4 1 2 1 2 1 1 1 2 4 4 3 3 3 3 3 2 3 2 3 2 4 4 4 4 4 3 3 3 4 3 3 Samples III>15 ALG1 ALG2 ALG3 ALG4 ARZ1 ARZ2 BEJ1 BEJ2 BOU1 BOU2 FET1 FET2 JIJ1 JIJ2 ORA1 ORA2 HALG1 HALG2 HALG3 HALG4 HAN1 HAN2 HAN3 HAN4 HBEJ1 HBEJ2 HGZ1 HGZ2 HORA1 HORA2 A1 2 1 1 1 2 1 2 1 1 2 1 1 2 1 2 2 1 3 2 3 3 4 4 3 3 3 2 3 4 4 3 A2 2 1 1 1 2 1 1 1 1 1 1 1 1 1 1 1 1 3 2 3 3 4 3 3 3 3 2 3 3 4 2 A3 2 2 1 2 2 1 1 2 1 1 1 1 2 1 3 2 3 3 2 4 3 4 4 3 4 4 2 4 4 4 2 A4 3 1 1 2 2 2 2 1 1 1 2 1 2 1 1 3 1 3 2 4 3 4 3 3 3 3 3 3 3 4 3 22 Discussion As underlined by several authors (see Diaz et al., 2004; Devlin et al., 2007) there were dramatically increase of indices to assess the quality of benthic ecosystem. In some specific cases, such as in transitional waters as in estuary due to the “Paradox of estuarine quality” (Dauvin, 2007; Elliott and Quitino, 2007), this development appeared necessary. But in several other cases as suggested by Diaz et al (2004), there was an inflation of new proposed indices or some rearrangements of existing indices only for their adaptation to local or regional conditions. So, more or less complicate indices continued to be proposed and published (see Dauvin et al., 2010). Face to this reality, two mains questions could be discussed: 1) do it is necessary to use complicate indices to assess the quality of benthos, and 2) what are the respectively parts of subjectivity and objectivity in the assessment of benthic quality status. Consequently, the main aim of this paper was to test the ability of the several kinds of benthic indicators, ability of sentinel species and Best Professional Judgment (BPJ) to assess the quality of soft-bottom benthic communities. This exercise was prepared on a selection of dataset of benthic samples able to characterize several shallow and harbour unpolluted or polluted situations to propose answers to the underlined previous questions. Statistical tests showed that the three chosen indices, i.e. H’, AMBI, and BO2A, gave similar ECoQS. Nevertheless, both AMBI and BO2A indices gave often the same assessment for a lot of samples (Annexe). The ITI approach was based on different concept, and we do not recommended to use this index for benthic EcoQS. A lot of recent papers (Labrune et al., 2006; Blanchet et al., 2008; Borja and Dauer, 2008; Borja et al., 2008; Bakalem et al., 2009; Dauvin et al., 2009; Grémare et al., 2009; Ranasinghe et al., 2009) had highlighted the difference of assessments of diverse benthic indicators. This was due to several reasons among them the fact that indices do not take into account the same factor of classification of species, i.e. classification of species into Ecological Groups such as in AMBI, classification of species into Trophic Groups such as in ITI or the numbers of species and individuals in a sample as in the H’ Shannon diversity… Authors promoted also the use of numerous indices and multi-criteria approaches for a better assessment of Ecological Quality (Muxika et al., 2007; Blanchet et al., 2008; Borja et al., in press). Most of the benthic experts recognized that there was certain subjectivity to the classification of species into ecological and biological groups, and the need to encourage studies on the respective role of benthic species in the functioning of benthic ecosystem, i.e. what they efficiently consumed, what was their sensitivity or tolerance versus diverse pollutions (Dauvin et al., 2010). Moreover, Grémare et 23 al. (2009) underlined that the use of a single universal sensitivity/tolerance list for different geographical areas appeared inappropriate for benthic species that are able to adapt their behaviour to local or regional environmental conditions. This suggested capacity of species adaptation in relation to geographical and environmental conditions and probably to species interactions. This was incompatible with the concept of a worldwide index, at least for those using species level of identification in their computation. A majority of the benthic indices was the need to identify the individuals to the species level, but as underlined in several recent papers (Boreo, 2010), the number of taxonomists has decreased dramatically, and in the future, it will no doubt be difficult to correctly denominate all the species found in macrobenthic communities (Dauvin, 2005). Furthermore, environmental impact assessments are often conducted by research consultancies operated by generalists who do not have sufficient species expertise to correctly identify the fauna in the study sites. Nonetheless, they continue to establish extensive lists of species for those paying for the study (Dauvin, 2005). The use of the Taxonomic Sufficiency (TS) concept can reduce errors in the assignment of ecological quality status, especially if EG are used, as in AMBI (Dauvin et al., 2010). Both indices BOPA then BO2A (Dauvin and Ruellet, 2007, 2009) followed the TS concept and can be used in coastal and estuarine environments with success. BO2A, using only the ratio between amphipods and opportunistic annelids, was a good alternative when the level of identification of the species was doubtful. The dominance of sentinel species used in this study appeared not adequate to give an accurate assessment of soft-bottom communities; nevertheless, their presence and dominance served to the experts to their Professional Judgement (Table 5). Subsequently, sentinel species remained for the observer an important role of “warning” vis–à-vis possible unbalances in the surrounding environment or distortions in the functioning of a benthic community. In fact, opportunistic species were present in a large number of samples in low abundances, as species in reserve which could increase rapidly their populations when favourable (increase of organic matter) suddenly rose. This was the typical behaviour of opportunistic species with a ‘r’ demographic strategy. Conversely, the presence and the dominance of the sensitive species to hydrocarbons and organic matter such as amphipod Ampelisca informed on the good quality of benthic environment. Their absence was not sufficient itself to prove an affected community. Another handicap in large dataset of samples is that some of them showed very low species richness and/or very low number of individuals. Our own experiences showed that three main causes could be at the origin of the collection of poor samples. 1) Hydrodynamism or anoxic 24 conditions; sediment composition and quantity of organic matter in the sediment can permit to identify the effective cause. 2) Fauna is rare due to natural physical and chemical constraints or human constraints such as dredging in harbours or navigational channels. 3) The size of the sample was too small. As most of the indices on the macrobenthic fauna are based strictly on species richness and abundances of each species or/and the total numbers of species and individuals in a sample, we suggested that samples with abundance lower than 20 individuals were not taken into account in the evaluation. Subsequently, it was always difficult to give a diagnostic for such poor samples. Finally BPJ exercise made in this study confirmed the previous studies of Weisberg et al. (2008) and Teixeira et al. (2009) on the pertinence and accuracy of such approach. It was probably due to the same learning of the professional experts on benthos and their high appropriation of the Pearson and Rosenberg paradigm on response of benthic community to increase of organic matter. Our study on several kinds of sites affected by different disturbances permitted to illustrate general points: - Indices and BPJ, as opportunist sentinel species, were highly relevant in polluted harbours and large emissary like Cortiou to give an objective assessment probably in relation to low hydrodynamism and large discharge and storage of organic matter in the sediment. - Indices and BPJ were not sufficiently sensitive to identify any disturbance in the case of low discharges of organic matter or other sources of chronic pollutions, i.e. input of metals and organics contaminants in the coastal waters such as along the Castellon coast where the benthos appeared in High or Good conditions, and in the Bay and Seine estuary in spite of a very high contaminated environment (Dauvin et al., 2007). Only zones in the navigational channel under harbour dredging in low salinity were identify as disturbed. - The coastal Algerian bays and gulfs appeared in good ecological quality at the end of the 1980’s and the beginning of the 1990’, the sampling sites at this period could be used latter as reference point (Bakalem et al., 2009). The presence of high abundances of tolerant species in estuarine conditions makes difficult to differentiate polluted areas, like in the ‘Rivière de Morlaix’ site (Bay of Morlaix) where the polychaete Chaetozone gibber dominated in “normal conditions” hides the Amoco Cadiz oil spill effect. In fact, such transitional environments constituted variable ecosystems naturally organically rich and the biota thus is similar to anthropogenically-organic rich areas. This was the catalyst for the concept of ‘estuarine quality paradox’ (Dauvin, 2007; Elliott and Quintino, 2007): ‘the dominant estuarine community is adapted to and reflects the high spatial and 25 temporal variability of highly naturally-stressed areas. However, this community has features very similar to those found in anthropogenically-stressed areas, thus making it difficult to detect anthropogenically-induced stress in estuaries. Furthermore, as estuaries are naturally organically rich, the biota thus is similar to anthropogenically-organic rich areas. Because of this, there is a danger that any indices based on these features and used to plan environmental improvements will be flawed. - The existing indicators based on classification of species in Ecological Groups such as AMBI were not susceptible to identify the effects of oil spills on benthic communities. In both oil spills, respectively the Amoco Cadiz in the Northern coast of Brittany, France and the Ares-Betanzos Ria, north Galicia Spain, the main effects on soft-bottom fine muddy fine sand communities were the disappearance of the sensitive amphipods species to hydrocarbon toxicity especially the species of the genus Ampelisca (Dauvin, 1998; Gomez Gesteira and Dauvin, 2005). Just after the spill, some Ampelisca specimens survived in the communities which showed high species richness and abundances; they were very restricted increases of opportunistic species except in the Bay of Morlaix with the opportunistic polychaeta Pseudopolydora pulchra proliferated in 1982 four years after the spill. Moreover, this peak of abundance was identify only during a very short summer period, demonstrating that the survey after an oil spill must be designed with a adequate high year frequency (monthly or quarterly frequency) and during a sufficient long time after a spill. The recovery time of softbottom communities after an oil spill was about 7-10 years (Dauvin, 1998, 2010). - Shannon H’ diversity index was not relevant to detect stressed or polluted environments. It was too sensitive to the species richness and to the total abundances in the samples. In the case of the Pierre Noire site, H’ was relatively low before the spill (2.2-2.3) corresponding to a Moderate status), then increased just after the spill to reach 3.7-4.9 values (Good and High status) and decrease dramatically during the summer peak of the polychaete Pseudopolydora pulchra (1.13 in August 1982, Poor Status, annexe 2). At Rivière de Morlaix H’ remained feeble due to the dominance of the tolerant Chaetozone gibber species, while in the AresBetanzos Ria, north Galicia Spain H’ showed always elevated values (3.8-5.4) corresponding to Good and High Status. The question of the Ecological Quality of benthic community was associated to the establishment of an Ecological Quality Ratio, i.e. the ratio between the quality observed in a site and a reference site. This concept of reference site was largely discussed (Borja and Muxika, 2005; Muxika et al., 2007), and not easy to solve especially in stressed environment such as harbours and transitional waters. Often, the ‘best’ sample in terms of species richness 26 and Shannon H’ was selected as reference (Grimes et al., in press). It was unmistakable that observations before a stress such as in the case of the Amoco Cadiz oil spill (Dauvin, 1998, 2000) permit to measure the effect of a disturbance. The accumulation of data along the North-Atlantic coast and western Mediterranean Sea such as along the Algerian coats (Bakalem et al., 2009; Grimes et al., in press) guaranteed the knowledge of the benthic status at a certain period but could not necessary be the reference status without human activities. This point underlined the fact that most of the coastal and transitional waters were under the anthropogenic influences more than one century and that it was very difficult and probably impossible to find any km of marine coast without man effect. Long-term survey of some benthic sites could also guarantee very interesting information about the resistance, resilience and to measure the range of the climatic natural fluctuations of coastal ecosystems. Such data were often necessary to measure with precision the impact of pollution (Dauvin, 2010). Macrobenthos being an excellent candidate for determining the Ecological Quality Status of the surrounding water masses, we suggested in the future, some practical recommendations mainly at the address of the research consultancies which are in charge of the assessment of the benthic quality in numerous impact studies: - The consultation of databases existing in an area such as the MacroBen base (Grémare et al., 2009) to identify the range of the demographic benthic parameters (species richness, abundance, dominant species) in contrasted areas: open zone, bays and transitional zones with the human impact was low. - The request if experts were available and implicated in benthic quality assessment, their Best Professional Judgement as a first approach. - To identify the taxa at least at a family level, which permit to assess the Benthic Quality Status at least on two types of indices: those based on Ecological Groups of family, such as AMBI, or BO2A, because it is easy to use and is rooted in the Sufficient Taxonomic principle; those based on the diversity, such as the Shannon diversity index H’ which showed moderate changes when it was calculated at a species, genus or family levels (Dauvin et al., 2009). We do not recommend the use of the ITI for benthic quality assessment. In conclusion, it appears that biotic indices privileged the identification of species resistant and tolerant to anthropogenic disturbance. Whether developed "objectively" or "subjectively", the indices that followed served only to reinforce this disturbance phenomenon, which contributed to highlighting the species that resisted the disturbances (Dauvin et al., 2010). The existing approaches to state on the ECOQS remained perfectible, including subjective ones. We encourage in the future pragmatic and simple methods to the research consultancies that 27 are mostly responsible for assessing benthic quality in numerous impact studies. When, we want to categorize the EcoQS of a benthic community the question is neither to identify the biodiversity of this community, nor to furnish elements on the benthic ecosystem functioning. Clearly, best professional judgement and taxonomic sufficiency should be encouraged in such diagnostic approaches. Acknowledgment The authors thank all the persons who had participated to the collection and sorting benthic invertebrates and Dr. Angel Borja (AZTI Tecnalia) for providing access to the Excel software sheet used to compute the Kappa analysis results. This work is a part of Jean-Claude Dauvin's contribution on benthic indices for the BEEST LITEAU project, funded by the French Ministry of Ecology, Energy, Sustainable Development and Sea (MEEDDM) and the three regional Water Agencies: Seine Normandie, Bretagne-Pays de Loire and Adour Garonne. References Bakalem, A., 2008. Les peuplements des sables fins de la côte Algérienne. Thèse de Doctorat d’Etat, Université des sciences et de la technologie, Houari Boumediene, faculté des sciences biologiques, Algérie, 677 pp. Bakalem, A., Ruellet, T., Dauvin, J.C., 2009. Benthic indices and ecological quality of shallow Algeria fine sand community. Ecological Indicators 9, 395-408. Bellan, G., 1967a. Pollution et peuplements benthiques sur substrat meuble dans la région de Marseille.Deuxième Partie. L'ensemble portuaire marseillais. Revue Internationale d’Océanographie Médicale, 8, 51-95. Bellan, G., 1967b. Pollution et peuplements benthiques sur substrat meuble dans la région de Marseille. Première Partie. Revue Internationale d’Océanographie Médicale, 6-7, 53-87. Bellan, G., Bourcier, M., Salen-Picard, C., Arnoux, A., Cesserley, S., 1999. Benthic Ecosystem Changes Associated with Wastewater Treatment at Marseilles: Implications for the Mediterranean Protection and Recovering. Water Environment Research, 71, 483-493. Bellan, G., 2008. Pollution indices. Encyclopedia of Ecology, 2861-2868. Blanchet, H., Lavesque, N., Ruellet, T., Dauvin, J.C., Sauriau, P.-G., Desroy, N., Desclaux, C., Leconte, M., Bachelet, G., Janson, A.L., Bessineton, C., Duhamel, S., Jourde, J., Mayot, S., Simon, S., De Montaudouin X., 2008. Use of biotic indices in semi-enclosed coastal 28 ecosystems and transitional waters habitats - Implications for the implementation of the European Water Framework Directive. Ecological Indicators, 8, 360-372. Boero, F., 2010. The study of species in the Era of biodiversity: a tale of stupidity. Diversity 2, 115-126. Borja, A., Franco, J., Perez, V., 2000. A marine Biotic Index to establish the ecological quality of soft-bottom benthos within European estuarine and coastal environments. Marine Pollution Bulletin 40, 1100-1114. Borja, A., Muxika I., 2005. Guidelines for the use of AMBI (AZTI's Marine Biotic Index) in the assessment of the benthic ecological quality. Mar. Pollut. Bull. 50: 787-789. Borja, A., Josefson, A.B., Miles, A., Muxika, A., Olsgard, F., Phillips, G., Rodriguez, G., Rygg, B., 2007. An approach to the intercalibration of benthic ecological status assessment in the North Atlantic ecoregion, according to the European Water Framework Directive. Marine Pollution Bulletin 55, 42-52. Borja, A., Dauer, D. M., 2008. Assessing the environmental quality status in estuarine and coastal systems: Comparing methodologies and indices. Ecological Indicators 8, 331-337. Borja, A., Dauer, D., Diaz, R., Llansó, R.J., Muxika, I., Rodriguez, J.G., Schaffner, L., 2008. Assessing estuarine benthic quality conditions in Chesapeake Bay: A comparison of three indices. Ecological Indicators 8, 395-403. Borja, A., Basset, A., Bricker, S., Dauvin, J.C., Elliott, M., Trevor Harrison, T., Marques, J.C., Weisberg, S.B., Ron West, R., in press. Classifying Ecological Quality and Integrity of Estuaries. Treatise on Estuarine and Coastal Science. Chapter 1.9. Cohen, J., 1960. A coefficient of agreement for nominal scales. Educational and Psychological Measurement 20, 37–46. Dauvin J.C., 1998. The fine sand Abra alba community of the Bay of Morlaix twenty years after the Amoco Cadiz oil spill. Marine Pollution Bulletin, 36, 669-676. Dauvin, J.C., 2000. The muddy fine sand Abra alba-Melinna palmata community of the Bay of Morlaix twenty years after the Amoco Cadiz oil spill. Marine Pollution Bulletin, 40, 528536. Dauvin, J.C., 2005. Expertise in coastal zone environmental impact assessments. Marine Pollution Bulletin 50, 107-100. Dauvin, J.C., 2007. Paradox of estuarine quality: benthic indicators and indices, consensus or debate for the future. Marine Pollution Bulletin 55, 271-281. Dauvin, J.C., Ruellet, T., 2007. Polychaete/amphipod ratio revisited. Marine Pollution Bulletin 55, 215-224. 29 Dauvin, J.C., Ruellet, T., Desroy, N., Janson, A.L., 2007. The ecological quality status of the Bay of Seine and the Seine estuary: use of biotic indices. Marine Pollution Bulletin 55, 241257. Dauvin, J.C., Ruellet, T., 2009. The Estuarine Quality Paradox: is it possible to define an Ecological Quality Status for specific modified and naturally stressed estuarine ecosystems? Marine Pollution Bulletin 59, 38-47. Dauvin, J.C., Bachelet, G, Barillé, A.L., Blanchet, H., De Montaudouin, X., Lavesque, N., Ruellet, T., 2009. Development of benthic indicators and index approaches in the three main estuaries along the French Atlantic coast (Seine, Loire and Gironde) for the implementation of the European Water Framework Directive (WFD). Marine Ecology, an evolutionary perspective 30, 228-240. Dauvin, J.C., Bellan, G., Bellan-Santini, D., 2010. Benthic Indicators: from subjectivity to objectivity – Where is the line? Marine Pollution Bulletin, 60, 947-953. Dauvin, J.C., 2010. Benthic time-series observations in North-eastern European Marine Station: Is a European Label Needed? Marine Pollution Bulletin 60, 483-488. De-la-Ossa-Carretero, J.A., Del Pilar Ruso, Y., Giménez Casalduero, F., Sánchez Lizaso, J.L., 2008. Effect of Sewage Discharge in Spisula subtruncata (da Costa 1778) Populations. Archives of Environmental Contamination and Toxicology 54, 226-235. De-la-Ossa-Carretero, J.A., Del-Pilar-Ruso, Y., Giménez-Casalduero F., Sánchez-Lizaso, J.L., 2009. Testing BOPA index in sewage affected soft-bottom communities in the northwestern Mediterranean. Marine Pollution Bulletin 58, 332-340. De-la-Ossa-Carretero, J.A., Del-Pilar-Ruso, Y., Giménez-Casalduero F., Sánchez-Lizaso, J.L., 2010. Sensitivity of tanaid Apseudes latreillei (Milne-Edwards) populations to sewage pollution. Marine Environmental Research 69, 309-317. De-la-Ossa-Carretero, J.A., Dauvin, J.C. submitted. A Comparison of Two Biotic Indices, AMBI and BOPA/BO2A, for Assessing the Ecological Quality Status (EcoQS) of Benthic Communities. Transitional Water Bulletin. Devlin, M., Best, M., Haynes, D., 2007. Implementation of the Water Framework Directive in European marine waters. Marine Pollution Bulletin 55, pp. 297. Diaz, R. J., Solan, M., Valente, R. M., 2004. A review of approaches for classifying benthic habitats and evaluating habitat quality. Journal of Environmental Management 73, 165-181. Elliott, M., Quintino, V.M., 2007. The Estuarine Quality Paradox, Environmental Homeostasis and the difficulty of detecting anthropogenic stress in naturally stressed areas. Marine Pollution Bulletin 54, 640-645. 30 Fleiss, J.L., Cohen, J., 1973. The equivalence of weighted Kappa and the intraclass correlation coefficient as measures of reliability. Educational and Psychological Measurement, 33, 613–619. Gomez Gesteira, J.L, Dauvin, J.C., 2005. Impact of the Aegean Sea oil spill on the Ría of Ares and Betanzos fine sand community (Northwest Spain). Marine Environmental Research, 60, 289-316. Grémare A., Labrune C., Vanden Berghe E., Amouroux J. M., Bachelet G., Zettler M. L., Vanaverbeke J., Fleischer D., Bigot L., Maire O., Deflandre B., Craeymeersch J., Degraer S., Dounas C., Duineveld G., Heip C., Herrmann M., Hummel H., Karakassis I., Kedra M., Kendall M., Kingston P., Laudien J., Occhipinti-Ambrogi A., Rachor E., Sardá R., Speybroeck J., Van Hoey G., Vincx M., Whomersley P., Willems W., Wlodarska-Kowalczuk M., Zenetos A., 2009. Comparison of the performances of two biotic indices based on the MacroBen database. Marine Ecology Progress Series, 382, 297–311. Grimes, S., 2003. Bilan et Diagnostic National de la pollution marine de la côte algérienne liée à des activités menées à terre. Programme d’actions stratégiques (PAS) destiné à combattre la pollution due à des activités menées à terre et de sa stratégie opérationnelle. Final Report PAM/PAS MED/MEDPOL,119 pp. Grimes, S., Ruellet, T., Dauvin, J.C., in press. Ecological Quality Status of the soft-bottom communities from the Algeria coast: general patterns and diagnostic. Marine Pollution Bulletin. Grimes, S., in préparation. Les peuplements macrobenthiques des substrats meubles algériens : organisation et structure. Thèse de doctorat d’état. Université Es Sénia, Oran. ICES, 2008. Report of the Workshop on Benthos Related Environment Metrics (WKBEMET), 11–14 February 2008, Oostende, Belgium. ICESCM2008/MHC:01. 53 pp. Labrune, C., Amouroux, J.M., Sarda, R., Dutrieux, E., Thorin, S., Rosenberg, R., Grémare, A., 2006. Characterization of the ecological quality of the coastal Gulf of Lions (NW Mediterranean). A comparative approach based on three biotic indices. Marine Pollution Bulletin, 52, 34-47. Landis, J.R, Kosch, G.G., 1977. The measurement of observer agreement for categorical data. Biometrics 33, 159–174. Mearns, A.J., Word, J.Q., 1982. Forecasting effects of sewage solids on marine benthic communities. In: Mayer G.F. (Ed.), Ecological stress and the New York Bight: science and management. Columbia, Estuarine Research Federation, pp. 495–512. 31 METAP, 1993. Etude de protection contre la pollution des ports et du littoral algériens. Rapport partiel 2 : détermination de l’origine de la pollution des ports et rades portuaires et au voisinage des principales plages algériennes. Institut pour la Qualité Des eaux (Danemark), Laboratoire d’Etudes Maritimes (Algérie) et Institut Danois d’Hydraulique (Danemark). Rapport de juin 1993, 35 pp. METAP, 1994. Etude de protection contre la pollution des ports et du littoral algériens. Rapport final. Institut pour la Qualité Des eaux (Danemark), Laboratoire d’Etudes Maritimes (Algérie) et Institut Danois d’Hydraulique (Danemark), 166 pp. Monserud, R, Leemans, R., 1992. Comparing global vegetation maps with the Kappa statistic. Ecological Modelling, 62, 275–293. 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. Marine Pollution Bulletin 55, 16-29. Pearson, T. H., Rosenberg, R., 1978. Macrobenthic succession in relation to organic enrichment and pollution of the marine environment. Oceanography and Marine Biology: An Annual Review 16, 229-311. Pinto, R., Patricio, J., Baeta, A., Fath, B. D., Neto, J. M., Marques, J. C., 2009. Review and evaluation of estuarine biotic indices to assess benthic condition. Ecological Indicators 9, 125. Rakocinski, C.F., G.A. Zapfe, 2005. Macrobenthic Process-Indicators of estuarine condition. In: Bortone, S.A (Ed.), Estuarine indicators. CRC Marine Science Series. CRC Press, pp. 316-331. Ranasinghe, J. A., Weisberg, S. B., Smith, R. W., Montagne, D. E., Thompson, B., Oakden, J. M., Huff, D. D., Cadien, D. B., Velarde, R. G., Ritter, K. J., 2009. Calibration and evaluation of five indicators of benthic community condition in two California bay and estuary habitats. Marine Pollution Bulletin 59, 5-13. Reish, D.J., 1959. An ecological study of pollution in Los Angeles-Long Beach Harbors. California Publication occasional papers, 22, 1-117. Ruellet, T., Dauvin, J.C., 2007. Benthic Indicators: analysis of the threshold values of Ecological Quality classifications for transitional waters. Marine Pollution Bulletin 54, 17071714. Scherrer, B., 1984. Biostatistique. Gaëtan Morin, Paris. Shannon, C.E., 1948. A mathematical theory of communication. Bell Syst. Tech. J. 27, 379423. 32 Teixeira, H., Borja, A., Weisberg, S.B., Ranasinghe, J.A., Cadie D.B., Dauer, D.M., Dauvin, J.C., Degraer, S., Diaz, R.J., Grémare, A., Karakassi, I., Llanso, R.J., Lovell, L.L., Marques, J.C., Montagne, D.E., Occhipinti-Ambroggi, A., Rosenberg, R., Sarda, R., Schaffner, L.C., Velarde, R.G., 2010. Do experts across North America and Europe produce comparable assessments of marine benthic infaunal community condition using best professional judgement? Marine Pollution Bulletin 60, 598-600. Vincent, C., Heinrich, H., Edwards ,A., Nygaard, K., Haythornthwaite, J., 2002. Guidance on typology, classification and reference conditions for transitional and coastal waters. Commission Européenne, CIS WG 2.4 (COAST), 119 pp. Weisberg, S. B., Thompson, B., Ranasinghe, J. A., Montagne, D. E., Cadien, D. B., Dauer, D. M., Diener, D. R., Oliver, J. S., Reish, D. J., Velarde, R. G., Word, J. Q., 2008. The level of agreement among experts applying best professional judgment to assess the condition of benthic infaunal communities. Ecological Indicators 8, 389-394. 33 Annexe. Main characteristics of the 121 samples selected in the dataset; Amp: Ampelisca; Opp, opportunist species; A: absence, Dm: dominance in %; SR, Species Richness; N: number of individuals. Ecological quality status (EcoQS), for Shannon H’, AMBI, BO2A and ITI. H, High (bleu); G, Good (green); M, Moderated (yellow); P, Poor (orange) and B, Bad (red); N, Normal (bleu), P, Perurbbed (yellow), and D, Degraded (red). BS: Bay of Seine; RM: Rivière de Morlaix, Bay of Morlaix; PN: Pierre Noire, Bay of Morlaix; X and Z, Galicia, 02, February…; /77, 1977…; Castellon coast, Vinaroz (VIN), Benicarló (BEN), Peñíscola (BEN), Alcossebre (ALC) and Torreblanca (TOR), V1, near the emissary, N1, 200 m from the emissary, N2, 1000 m from the emissary; C, Cortiou; II, II/III, III<15, III> 15, Marseilles harbour; ALG, Algiers; ARZ, Arzew; BEJ, Bejaia; FET, Fetzara; JIJ, Jijel; OR, Oran; HALG, Algiers harbour; ; HAN, Annaba harbour; HBEJ, Bejaia harbour; HGZ Ghazaouet harbour; HORA, Oran harbour. Samples BS2 BS7 BS10 BS11 BS15 BS24 BS28 BS40 BS58 RM08/77 RM02/78 RM08/78 RM02/78 RM08/79 RM02/80 RM08/80 RM02/81 RM08/81 RM03/82 RM08/82 RM03/83 PN08/77 PN03/78 PN08/78 PN03/79 PN08/79 PN03/80 PN08/80 PN03/81 PN08/81 PN03/82 Amp. Dm 4,89 1,01 0,19 0,26 0,02 1,80 0,31 A A 0,41 0,62 A A A 0,44 1,07 0,13 0,32 0,11 0,61 0,17 85,11 84,06 0,17 0,48 2,71 0,91 4,64 7,48 12,09 31,38 Opp. Dm 0,22 0,38 A A 0,02 0,50 0,53 A A A A A A 0,03 A 0,01 0,04 0,02 0,13 0,01 0,11 0,01 A 0,11 A 0,33 0,02 A A 0,03 0,08 SR N H' AMBI BO2A ITI 75 61 37 43 57 85 86 12 5 34 45 47 52 55 39 72 56 66 63 81 49 92 81 81 61 86 86 133 81 115 88 10611 4252 1026 377 5570 4826 13444 84 52 4185 1943 6594 3930 3661 1571 6283 2364 4694 4449 12223 1753 29250 8681 1747 2084 2806 4502 4758 2492 3481 2460 G G M G P G M M M P M P M G M M G M M M G M M G G H G H H H G G G H H G G G H G M M M M M M M G M M M G H H G G G G G G G G M G H H H M G H H P M B P P P P M P P P G H H G G G G G G G G P P N P N P P P D P P P P P P P P P P P P N N P P P P P P P P Samples Amp. Dm PN08/82 3,20 PN03/83 40,44 X1292 8,16 X0393 A X0893 A X0394 A X0994 A X0495 0,78 X0895 2,23 X0496 A X0796 9,33 X1196 3,55 Z1192 A Z0393 A Z0893 A Z0394 A Z0994 A Z0495 A Z0895 2,13 Z0496 0,75 Z0796 7,89 Z1196 1,47 VIN.V1 8,84 VIN.N1 7,27 VIN.N2 18,31 BEN.V1 2,53 BEN.N1 5,83 BEN.N2 0,94 PEN.V 1,05 PEN.N1 3,26 PEN.N2 3,75 Opp. Dm 0,05 A A 3,51 A A A A A A 0,22 0,25 A 1,11 0,44 0,77 A A A A A A 20,41 1,09 0,45 6,32 3,58 0,40 17,39 0,81 1,50 SR N H' AMBI BO2A ITI 133 89 54 26 32 29 33 40 59 51 58 62 33 33 40 24 43 37 54 40 55 62 42 41 55 46 45 40 30 35 37 37658 2465 249 114 106 85 71 129 404 337 450 394 108 91 228 130 298 111 282 134 317 204 3675 6875 10975 3950 5575 7475 4600 6125 3325 P G H H H G H H H H H H H H G G H H H H H H H H H H H G G G H M H G G G G H G G G G G H G G H H G G G G G G G H G G H G H G B H H G H G H H G H G G H G H H G G G H H G G H H G G H G G G N P P P P P P P P N P P P P P P P P P P P P P P P P P P P N P 34 Samples ALC.V1 ALC.N1 ALC.N2 TOR.V1 TOR.N1 TOR.N2 C65C6 C89I7e C94I7e C95I7e C96I7e C01I7e C76F2 C76E6 C76D2 C94F2 C94E6 C94D2 II II II/III II/III II/III III<15 III<15 III<15 III>15 III>15 III>15 ALG1 ALG2 Amp. Opp. Dm Dm 13,04 A 1,77 1,32 6,42 0,40 20,45 0,75 1,16 0,58 0,68 5,40 A 100,00 A 100,00 A 2,70 A 40,81 A 24,32 A 10,92 0,59 2,85 2,80 A 2,43 0,61 A A 5,50 A 3,39 A A 100,00 A 98,21 A 96,91 A 88,96 A 45,25 A 10,49 A 2,04 A 60,00 A 7,30 A 3,03 A 22,72 1,33 A 2,86 0,26 SR N H' AMBI BO2A ITI Samples 33 40 44 35 36 38 3 3 15 22 8 27 93 105 84 46 44 26 3 7 9 17 11 13 29 11 24 16 20 45 77 3725 5650 6225 3300 4325 3700 8742 45 37 98 37 229 1001 710 327 147 108 59 114 448 422 444 179 162 978 105 370 132 142 748 1920 H G H H G H B P G G M H H H H H H H P P P M M M M M G M G M G H G H H G G B B H G G G G G G G G H B B B P G G G G G G G H G H H H H H G B B G M G M M G G G H G B B B M M M M G G G H H G P N P P N P N P P P N P P P P D P P P P P P P N D N D P P P P ALG3 ALG4 ARZ1 ARZ2 BEJ1 BEJ2 BOU1 BOU2 FET1 FET2 JIJ1 JIJ2 ORA1 ORA2 HALG1 HALG2 HALG3 HALG4 HAN1 HAN2 HAN3 HAN4 HBEJ1 HBEJ2 HGZ1 HGZ2 HORA1 HORA2 Amp. Dm A 0,64 A 2,85 5,06 9,04 33,30 5,07 2,56 6,06 27,65 A A A A A A A A A A A A A A A A A Opp. Dm 0,51 1,28 3,60 A A 0,56 A A A A 2,47 A 2,04 A 30,21 9,90 74,53 26,63 94,97 89,13 2,04 39,04 88,13 1,69 13,57 6,49 98,40 18,60 SR N H' AMBI BO2A ITI 45 67 59 21 63 45 13 69 36 27 49 11 11 28 46 31 11 25 13 4 31 22 25 39 19 8 5 17 786 1402 970 175 1580 354 195 1270 390 165 683 226 245 1745 383 101 106 214 338 92 1273 227 1230 2134 641 1032 502 43 G H H H M H G H H H H P P M H H M G P B P G P B P P B G G G G G H H H H H H H H G G G G P M B M M M M H M P B G G G H G H H H H G G G H H H M G M P B H G G G H B B B G N N P P N P N P P P P N D D P P P P P N P N N D P P P P