the use of the marine biotic index ambi in the assessment of the ecological status of the Óbidos...

www.elsevier.com/locate/marpolbul

Marine Pollution Bulletin 52 (2006) 1414–1424

The use of the marine biotic index AMBI in the assessment ofthe ecological status of the Obidos lagoon (Portugal)

Susana Carvalho a, Miguel B. Gaspar a,*, Ana Moura a, Carlos Vale b, Paulo Antunes b,Odete Gil b, Luıs Cancela da Fonseca c,d, Manuela Falcao a

a Instituto Nacional de Investigacao Agraria e das Pescas (INIAP/IPIMAR), Centro Regional de Investigacao Pesqueira do Sul (CRIPSul),

Av. 5 de Outubro s/n, 8700-305 Olhao, Portugalb Instituto Nacional de Investigacao Agraria e das Pescas (INIAP/IPIMAR), Av. Brasılia s/n, 1449-006 Lisboa, Portugal

c IMAR, Laboratorio Marıtimo da Guia, Estrada do Guincho, 2750-642 Cascais, Portugald FCMA, Universidade do Algarve, Campus de Gambelas, 8005-139 Faro, Portugal

Abstract

This study reports the longitudinal distribution of sediment properties, including inorganic and organic contaminants, and the struc-ture of the benthic community in Obidos lagoon, a coastal system permanently connected to the sea and with negligible freshwatersources. Sediments from the upper to central lagoon consist of fine particles (91%) and from the lower lagoon of sands (94%). Chemicalcomposition is strongly correlated to the percentage of fine particles. Contamination is relatively low in those sediments suggesting theeffect of diffuse sources. The increase in organic matter content from down- to upstream areas was associated with the dominance ofopportunistic species, while sensitive and indifferent species to organic enrichment were mainly associated to the clean sandy downstreamarea. The marine biotic index (AMBI) was suitable for the discrimination of the biological and environmental gradients in the Obidoslagoon and was highly related with the gradient of organic matter content in this system.� 2006 Elsevier Ltd. All rights reserved.

Keywords: Coastal lagoons; Sediment; Contaminants; Macrobenthic communities; NE Atlantic; AMBI

1. Introduction

Benthic assemblages are broadly recognised as a propertool to describe ecological conditions of marine and brack-ish systems. Nevertheless, some shallow coastal areas ofEurope remain poorly documented in terms of generalcomposition, structure or ecology (Rueda et al., 2001).Infaunal communities are in direct contact with the sedi-ment, where organic matter and multiple contaminantsare accumulated, and also have an important role on thestructure and functioning of ecosystems (Rhoads et al.,1978; Chapman et al., 1987; Rees et al., 1990). Distributionpatterns of macrobenthic communities are usually associ-

0025-326X/$ - see front matter � 2006 Elsevier Ltd. All rights reserved.

doi:10.1016/j.marpolbul.2006.04.004

* Corresponding author. Tel.: +351 289 700500; fax: +351 289 700535.E-mail address: [email protected] (M.B. Gaspar).

ated with environmental variables, such as salinity, grain-size, organic carbon, and the presence of contaminants(see reviews by Gray, 1974; Snelgrove and Butman,1994). The influence of each of these variables is difficultto isolate due to the dynamics of water column and spatialvariability of sediment properties existing in coastal ecosys-tems. However, with the implementation of the WaterFramework Directive (WFD) it is crucial, in Europeanwaters, to distinguish natural variability from effects ofanthropogenic pressures on macrobenthic communities.A wide variety of procedures have been developed aimingto assess the effects of environmental changes on the eco-logical quality status using macrobenthic communities.These procedures are either related with measures of com-munity structure (species abundance, diversity and faunalcomposition) and/or community function, such as speciescolonization rates (Long et al., 1995). The WFD establishes

mailto:[email protected]

Fig. 1. The Obidos lagoon. Location of the sampling stations.

S. Carvalho et al. / Marine Pollution Bulletin 52 (2006) 1414–1424 1415

that surface waters should be divided into Water Bodiessubjected to monitoring in order to determine their Ecolog-ical Quality Status (EcoQ). This is done through BiologicalQuality Elements and Supporting Quality Elements. TheEcoQ classes range from High to Bad quality status. TheWFD demands EU member states to assess the EcoQ ininland surface waters, transitional (or estuarine) waters,coastal and ground-waters (Borja et al., 2004b). The defini-tion of water bodies – management units – reflects both thesignificant pressures and the uniformity of state. Accordingto the WFD, the EcoQ should be assessed upon the biolog-ical, hydrological, hydromorphological and physico-chem-ical quality elements, with particular relevance for thebiological components (Borja et al., 2004c). Phytoplank-ton, macroalgae and benthos should be analysed in bothcoastal and transitional waters, while fish must also betaken into account for the ecological classification of tran-sitional water bodies (Borja et al., 2004b). One method tobe used when assessing the EcoQ of transition water bodiesis the abundance and species composition of benthic mac-rofauna that can be complemented with the concept of sen-sitive and tolerant taxa (Rosenberg et al., 2004). Severalecological indices have also been proposed in order toassess the benthic quality status (Borja et al., 2000; Simbo-ura and Zenetos, 2002; Rosenberg et al., 2004) and some ofthem were already adapted in conformity with the classifi-cations recommended by the WFD. One of the indicesextensively applied in the last years with good results isthe Marine Biotic Index (AMBI). This index was developedby Borja et al. (2000) and is based on the assignment of acertain ecological behaviour to each taxon observed duringa survey. Five ecological behaviours were established(sensitive, indifferent, tolerant, second-order opportunistictaxa, and first-order opportunistic taxa), based on datafrom different works and also on the experience of theauthors. Borja et al. (2004a) established the equivalencebetween the values observed by AMBI and the ecologicalstatus’ classification of the WFD. Moreover, anotheradvantage of this index is the fact that it has been exten-sively tested for different regions along the European coastby different scientific teams (see Muxika et al., 2005, for areview).

The Obidos Lagoon is located in the west coast of Por-tugal. This is a shallow lagoon with a mean depth of 1 mand a wet area of 7 km2. It is permanently connected tothe sea, although the position and shape of the narrow inletchannel have changed naturally during the last decades.The inner part of the lagoon contains extensive inter-tidalsand flats and channels, ending in two branches orientedto NW and SW, respectively. Freshwater inputs are negli-gible and water and particles are discharged into thesebranches only during rainy periods. The influence of thetide extends to the entire lagoon, without pronouncedlongitudinal variation of salinity and vertical density strati-fication (Instituto Hidrografico, unpublished data). Thecomposition and structure of the benthic communitywithin the lagoon were studied in the 1980s (Quintino,

1988) and discussed in light of the morphology, majorcharacteristics of water column and sediment grain sizedistribution.

This paper reports the longitudinal distribution of sedi-ment properties, including trace elements and organic con-taminants, and benthic communities in this coastal lagoon.The aim of the present work was to assess the applicabilityof the AMBI index in classifying the ecological qualitystatus of this system and to provide baseline informationconcerning the EcoQ for future monitoring studies.

2. Materials and methods

2.1. Sampling

Surface sediments were collected with a Van Veen grab(0.05 m2) at 24 stations located from upper to lower part ofthe Obidos Lagoon in February 2001 (Fig. 1). Five sampleswere collected at each station. Sediments of three replicateswere sieved on board through a 1 mm square mesh and theretained material was fixed in 4% buffered formalin stainedwith Rose Bengal for macrofaunal studies. The upper 1-cmlayer from a fourth replicate was sub-sampled for the deter-mination of organic carbon, chlorophyll a, phaeopigments,major, minor and trace elements, and organochlorine com-pounds. Samples were frozen until subsequent analysis.Additionally, a grab sediment sample was collected at eachstation for grain-size analysis.

1416 S. Carvalho et al. / Marine Pollution Bulletin 52 (2006) 1414–1424

2.2. Grain size and chemical analyses

The sediment particle size analysis was performed using63 lm and 2 mm sieves to separate the silt-clay, sand andgravel fractions, respectively. Each fraction was dried andweighed and sediments classified according to the percentageof each fraction (Moita, 1985). The upper sediment sampleswere dried to 80 �C to a constant weight and the ground to afine powder. Organic carbon (Corg) was determined with aCHN analyser as the difference between contents of totalcarbon and carbon after sediment being ignited to 450 �C.Acetanilide was used as reference material (Byers et al.,1978). The chlorophyll a (chl a) and phaeopigments (phaeo)were extracted for 24 h with 90% acetone from the upper wet

Fig. 2. Distribution of percentage of fine particles and total organic carbon (Co

from the Obidos lagoon.

sediments and determined by spectrophotometry accordingto Lorenzen equations (Plante-Cuny, 1974). For metalsdetermination sediments were mineralised by a mixture ofacids at 100 �C for 1 h (Rantala and Loring, 1977). Concen-trations of Al, Si, Ca, Mg, Fe, Mn and Zn were determinedby flame atomic absorption spectrometry and Cu, Pb, Cr,Ni, As and Cd determined by graphite furnace atomicabsorption spectrometer with Zeeman correction. Mercurywas measured by atomic absorption spectrometry in a mer-cury analyser LECO AMA-254 (Branco et al., 2004). Inter-national certified standards were used to ensure accuracyand precision was determined by analysing replicate sam-ples. Obtained and certified values did not differ signifi-cantly. Organochlorine compounds were analysed by gas

rg), chlorophyll a (lg g�1) and phaeopigments (lg g�1) in surface sediments


chromagraphy after extraction with hexane/acetone (1:1)during 18 h following the method described in Gil and Vale(2001). A mixture of 18 PCB congeners (IUPAC Nos 18, 26,52, 49, 44, 101, 151, 149, 118, 153, 105, 138, 187, 183, 128,180, 170, 194), p,p 0-DDT and metabolites and dieldrin wasused as external standard for quantification. Levels of DDTsand PCBs were only quantified in selected sand and mudsamples (stations 1, 9, 11, 13–16, 18, 20, 21, and 24), becausethe Obidos Lagoon is a nearly closed system without directemissions from industries, and freshwater inputs occurs inshort raining periods.

2.3. Benthos and data analysis

Samples for the study of macrobenthic communitieswere hand sorted into major taxonomic groups, identifiedto the lowest practical taxonomic level (usually specieslevel) and counted. Macrobenthic community structurewas analysed regarding the univariate variables, abundance(N), number of species (S) and diversity (Shannon–Wienerand Margalef’s species richness) and eveness (Pielou) indi-ces. Multivariate analysis was performed using PRIMERv5.0 software package (Clarke and Gorley, 2001). Similar-ity between stations was analysed by cluster techniques onthe biological matrix of 101 taxa · 24 sites after log (x + 1)transformation of the abundance values. The cluster anal-ysis applied the unweighted pair group average algorithmand the Bray–Curtis similarity coefficient. The assignmentof the identified species into one of the five ecologicalgroups proposed by the marine biotic index AMBI wasbased upon the list available in the AMBI program(http://www.azti.es). The ecological quality status of thesampled stations was determined using this biotic index.Correlations between biotic and abiotic variables were per-formed using STATISTICA v6. Organic compounds were

Fig. 3. Distribution of As (lg g�1) and tDDT (ng g�1) conc

not included due to the inexistence of values for allstations.

3. Results

3.1. Environmental parameters

In all the samples the percentages of sand and fine par-ticles (<63 lm) account for almost 100% and the gravelfraction was negligible. The grain size distribution indicatesthat Obidos Lagoon may be divided into two geographicalzones: zone I, including stations located in the innerbranches and central part, where sediments contain91% ± 11 of fine particles (stations 11–18, 20, 21) and theseaward zone II, corresponding to stations in channelsand inter-tidal flats consisting of 94% ± 10 of sand (sta-tions 1–10, 23, 24). The contrasting geographic distributionof the fine fraction of sediments is presented in Fig. 2.

The Al concentrations were highly correlated to thepercentage of fine particles (r2 = 0.96), indicating that thiselement represents well the contrasting presence of the finefraction of sediments in the lagoon. The correlationbetween Corg contents and fine fraction is less strong(r2 = 0.82) because sands from the central part of thelagoon containing considerable quantities of organic mat-ter, presumably plant and macroalgae litter transportedfrom upstream (Fig. 2). Chlorophyll a was not correlatedto Corg and concentrations did not vary significantly(p > 0.05) between sand and mud from the two zones.Higher quantities of phaeopigments were found in muddysamples, presumably reflecting the decomposition of Ulva

sp. that seasonally covers the upper areas of the lagoon.Concentrations of major, minor and trace elements werehighly correlated to values of percentage of fine fraction,Al and Corg (r2 > 0.81). Poorer correlations were found

entrations in surface sediments from the Obidos lagoon.

http://www.azti.es

Table 1Percentage of sand and fine particles (<63 lm) and concentrations ofmajor-, minor- and trace-element, organic carbon, chlorophyll a, phae-opigments, and organic contaminants (PCB, DDT and dieldrin) for thetwo zones established

Parameter Zone I Zone II p

Sand (%) 9 ± 11 94 ± 10 <10�10

Fine particles (%) 91 ± 11 6 ± 10 <10�10

Corg (%) 1.38 ± 0.22 0.17 ± 0.23 <10�10

Al (%) 10.9 ± 1.0 0.9 ± 0.57 <10�10

Si (%) 24.4 ± 1.3 39.2 ± 1.2 <10�10

Ca (%) 1.7 ± 0.3 2.0 ± 0.6 *

Mg (%) 1.78 ± 0.18 0.14 ± 0.05 <10�10

Fe (%) 5.4 ± 0.5 0.2 ± 0.1 <10�10

Mn (lg g�1) 410 ± 46 61 ± 38 <10�10

Zn (lg g�1) 128 ± 21 14 ± 1 <10�10

Cu (lg g�1) 56.5 ± 5.0 3.8 ± 0.2 <10�10

Pb (lg g�1) 45 ± 7 6 ± 3 <10�10

Cr (lg g�1) 117 ± 11 2.8 ± 2.2 <10�10

Ni (lg g�1) 79 ± 19 13 ± 2 <10�10

As (lg g�1) 22.2 ± 2.4 <5.0 <10�10

Cd (lg g�1) 0.152 ± 0.040 0.014 ± 0.002 <10�10

Hg (lg g�1) 0.10 ± 0.04 <0.06 <0.002Chlorophyll a (lg g�1) 1.7 ± 1.2 1.4 ± 1.1 *

Phaeopigments (lg g�1) 20 ± 24 1.5 ± 2.7 <10�10

tPCB (ng g�1) 1.2 ± 0.4 0.14 ± 0.15 <0.001tDDT (ng g�1) 15.5 ± 2.9 0.06 ± 0.08 <10�5

Dieldrin (ng g�1) 0.7 ± 0.3 <0.01 <0.02

p: Statistical significance.*p > 0.05.


for Ca that was relatively uniform in all sediments and forHg with concentrations being below detection limit in

Table 2Univariate variables, mean number of taxa (S) and mean abundance (N), and

Station S N Ecological groups (

(taxa/0.05 m2) (ind./0.05 m2) I

1 2.3 ± 1.53 9.3 ± 4.62 88.9 ± 12.73

2 0.7 ± 0.58 1.0 ± 1.00 100.0 ± 0.00

3 17.0 ± 0.00 412.3 ± 39.80 5.7 ± 2.874 1.0 ± 0.00 1.0 ± 1.00 0.0 ± 0.005 3.7 ± 0.58 5.3 ± 1.15 69.4 ± 17.35

6 4.7 ± 0.58 7.7 ± 1.53 46.3 ± 22.45

7 14.7 ± 4.93 79.0 ± 66.54 17.5 ± 4.768 12.3 ± .15 415.3 ± 83.81 1.9 ± 0.509 5.3 ± .53 9.7 ± 8.08 17.0 ± 5.13

10 11.0 ± 2.00 238.0 ± 99.86 5.3 ± 1.5411 11.0 ± .00 131.0 ± 38.59 0.0 ± 0.0012 10.3 ± 1.53 123.3 ± 31.09 0.4 ± 0.6413 10.0 ± 0.00 121.0 ± 21.66 0.5 ± 0.4414 9.7 ± 0.58 567.0 ± 145.35 0.0 ± 0.0815 8.7 ± 3.78 87.3 ± 76.35 3.6 ± 3.4616 15.7 ± 2.08 450.7 ± 153.63 2.0 ± 1.9117 14.0 ± 4.36 661.0 ± 102.19 0.5 ± 0.3218 13.3 ± 0.58 381.0 ± 183.74 0.0 ± 0.0019 13.3 ± 1.15 129.7 ± 29.14 1.4 ± 1.6320 16.0 ± 2.64 215.0 ± 68.55 2.5 ± 2.9421 10.0 ± 1.00 68.7 ± 61.16 2.0 ± 1.8422 11.7 ± 2.31 77.7 ± 28.29 0.0 ± 0.0023 12.0 ± 1.00 156.3 ± 41.43 0.5 ± 0.4724 14.3 ± 2.52 74.3 ± 43.47 19.5 ± 10.60

Ecological groups correspond to: I. sensitive to pollution; II. indifferent to pofirst-order opportunistic taxa (see Borja et al., 2000). In bold is represented th

sands. Organochlorines (PCBs, DDTs and dieldrin) wereonly determined in a selected number of samples (11) andthus poor correlations with interpretative parameters werefound. Fig. 3 illustrates the dramatic differences of DDTsand As concentrations between the two zones.

The concentrations of most of the analysed parametersdiffer significantly from zone I to zone II. On the basis ofthose differences it was calculated the mean compositionof sediments from the two zones (Table 1), excluding datafrom stations 19 and 22 since fine fraction and Al indicatean intermediate situation. The contrast was much morepronounced for chemical elements than for organochlo-rines. Only Ca and chlorophyll a concentrations showedmean values that did not differ significantly between thetwo zones.

3.2. Macrobenthic communities

A total of 13,268 individuals distributed among 101 taxawere identified in the lagoon. Among the major taxonomicgroups, only Polychaeta and Bivalvia presented a wide-spread distribution, although gastropods, oligochaetes,phoronids and isopods were also important groups. Sta-tions 1, 2 and 5 (located near to the sea inlet) are, ingeneral, dominated by the archiannelid Saccocirrus papillo-

cercus, the polychaete Scoloplos armiger and by the bivalveTellina tenuis. Molluscs and polychaetes co-dominated atstations 3, 4, 6–10, 23 and 24. Stations 11–22 located fromthe central to inner branch of the lagoon were mainly

relative percentage of the five ecological groups for each sampling station

%)

II III IV V

8.3 ± 8.33 2.8 ± 4.81 0.0 ± 0.00 0.0 ± 0.000.0 ± 0.00 0.0 ± 0.00 0.0 ± 0.00 0.0 ± 0.000.4 ± 0.49 90.2 ± 3.67 3.3 ± 1.91 0.4 ± 0.24

50.0 ± 70.71 50.0 ± 70.71 0.0 ± 0.00 0.0 ± 0.0030.6 ± 17.35 0.0 ± 0.00 0.0 ± 0.00 0.0 ± 0.0042.1 ± 11.81 7.4 ± 12.83 4.2 ± 7.22 0.0 ± 0.0011.8 ± 8.57 64.4 ± 10.68 6.2 ± 4.19 0.0 ± 0.000.5 ± 0.28 93.1 ± 5.10 2.9 ± 2.79 1.6 ± 1.98

52.2 ± 32.38 30.7 ± 37.29 0.0 ± 0.00 0.0 ± 0.002.7 ± 0.96 91.6 ± 0.95 0.4 ± 0.47 0.0 ± 0.00

13.2 ± 8.80 28.8 ± 6.41 38.0 ± 15.82 20.0 ± 4.135.7 ± 4.81 37.7 ± 4.37 44.1 ± 3.99 12.2 ± 4.97

26.6 ± 7.51 26.8 ± 8.78 45.7 ± 0.66 0.5 ± 0.821.3 ± 0.87 25.6 ± 0.88 63.0 ± 7.93 10.1 ± 7.767.1 ± 12.28 28.2 ± 23.9 16.7 ± 9.65 44.4 ± 26.12

43.1 ± 17.63 8.4 ± 2.33 44.9 ± 15.29 1.7 ± 1.3838.3 ± 6.90 12.9 ± 0.60 47.8 ± 6.19 0.4 ± 0.621.8 ± 0.63 25.8 ± 9.53 65.7 ± 4.54 6.6 ± 6.273.8 ± 1.25 25.2 ± 4.77 48.5 ± 12.11 21.1 ± 8.071.9 ± 0.57 11.4 ± 4.48 69.7 ± 11.99 14.4 ± 5.830.0 ± 0.00 39.6 ± 16.70 9.0 ± 4.76 49.4 ± 20.114.8 ± 5.72 48.6 ± 5.21 15.7 ± 7.47 31.0 ± 15.610.0 ± 0.00 68.8 ± 17.00 27.9 ± 16.28 2.8 ± 1.678.9 ± 6.13 64.9 ± 17.79 3.5 ± 4.58 3.2 ± 2.97

llution; III. tolerant to pollution; IV. second-order opportunistic taxa; V.e dominant ecological group for each station.

Table 3Mean and cumulative (Cum) abundance (Abd) of the three most abundant taxa at each station

Mean Abd Cum Abd Mean Abd Cum Abd

S1 S13

Saccocirrus papillocercus 8.0 85.7 Heteromastus filiformis 53.3 44.1Hesionura sp. 0.7 92.8 Phoronis spp. 33.0 71.4Hydrobia ulvae 0.3 96.4 Abra tenuis 15.7 84.3

S2 S14

Saccocirrus papillocercus 1.0 100.0 Heteromastus filiformis 350.3 61.8Abra tenuis 95.3 78.6Und. Oligochaeta 63.7 89.8

S3 S15

Hydrobia ulvae 275.3 66.6 Hydrobia ulvae 24.3 27.9Notomastus sp. 66.0 82.8 Capitella spp. 18.7 49.2Tapes decussatus 15.0 86.4 Und. Anthozoa 12.0 63.0

S4 S16

Microphthalmus similis 0.7 66.7 Phoronis spp. 179.3 39.8Hydrobia ulvae 0.3 100.0 Und. Insecta 122.0 66.9

Corbula gibba 86.0 85.9

S5 S17

Scoloplos armiger 1.7 31.3 Phoronis spp. 249.0 37.7Tellina tenuis 1.0 50.0 Corbula gibba 208.7 69.2Saccocirrus papillocercus 1.0 68.7 Heteromastus filiformis 101.3 84.6

S6 S18

Nephtys cirrosa 2.3 30.4 Heteromastus filiformis 252.0 66.1Ophelia neglecta 1.7 52.2 Und. Oligochaeta 32.0 74.5Tellina donacina 1.3 69.6 Cerastoderma edule 16.0 78.7

S7 S19

Hydrobia ulvae 27.0 34.2 Heteromastus filiformis 58.3 45.0Cerastoderma edule 15.7 54.0 Und. Oligochaeta 25.0 64.3Tellina tenuis 6.7 62.4 Cyathura carinata 16.7 77.1

S8 S20

Hydrobia ulvae 303.7 73.1 Heteromastus filiformis 146.0 67.9Abra ovata 46.3 84.2 Und. Oligochaeta 27.3 80.6Cerastoderma edule 30.3 91.5 Abra alba 8.3 84.5

S9 S21

Hydrobia ulvae 4.0 41.4 Capitella spp. 25.3 36.9Microphthalmus similis 1.0 51.7 Malacoceros fuliginosus 15.0 58.7Nephtys cirrosa 1.0 62.1 Notomastus sp. 8.7 71.4

S10 S22

Hydrobia ulvae 208.7 87.7 Notomastus sp. 30.3 39.1Scoloplos armiger 9.3 91.6 Capitella spp. 11.0 53.2Cerastoderma edule 7.3 94.7 Und. Oligochaeta 8.7 64.4

S11 S23

Heteromastus filiformis 45.3 34.6 Heteromastus filiformis 45.3 29.0Und. Oligochaeta 18.3 48.6 Abra ovata 40.7 55.0Phoronis spp. 17.3 61.8 Alkmaria romijni 24.7 70.8

S12 S24

Heteromastus filiformis 51.3 41.6 Und. Euclymeninae 20.3 27.4Scrobicularia plana 23.0 60.3 Hydrobia ulvae 16.0 48.9Und. Oligochaeta 16.0 73.2 Cerastoderma edule 9.3 61.4


dominated by annelids, except for stations 16 and 17 wherePhoronida, insect larvae and the bivalve Corbula gibba

were especially abundant. Spatial distributions of the meannumber of species and mean abundance are shown in Table2. In general, higher species values and abundance wereobserved in central and inner areas, while lower numberof species and mean abundance were found in the lower

zone of the lagoon. An exception was station 3 presenting28 species and 412 individuals. The mean percentage ofeach ecological group per sampling station is presentedin Table 2. Stations from the downstream area and navi-gation channel presented higher percentage of the ecologi-cal groups I (sensitive), II (indifferent) and III (tolerant),while in the inner area, percentage of the groups IV


(second-order opportunistic) and V (first-order opportunis-tic) were clearly higher. Only one station was dominated byfirst-order opportunistic taxa (station 21), while station 2only presented sensitive taxa. A transition zone wasobserved mainly in the stations located in the centralsand-bank, dominated by tolerant taxa (Table 2).

The relative and cumulative abundance per station forthe three more abundant taxa are presented in Table 3.For each sampling station the three most abundant speciesaccounted for more than 60% of the total abundance.

The cluster analysis of biological data identified 2 mainstation groups: A and B (Fig. 4). Group A comprised twosubgroups Group A1, with stations located in the naviga-tional channel (stations 4, 6 and 9) and Group A2, with sta-tions located near the lagoon inlet (stations 1, 2 and 5).Group B constitutes a main cluster including stations fromcentral to inner areas. From the cluster analysis it seemsthat Group B1, although associated to Groups B2–B4,constitutes a transition zone between inner and inlet areas,emphasized by its faunal composition as this group com-prises species that either occurs in the inner or inlet areas(Table 4). A thorough analysis on the faunal compositionof the faunistic groups was presented in a previous paper(Carvalho et al., 2005).

In the Obidos lagoon, AMBI values ranged from 0.0(station 2) to 4.6 (station 21) (Fig. 5). According to the clas-sification proposed by Borja et al. (2000), in terms of ben-thic community health this lagoon presented stationsclassified as normal (stations 1 and 2), impoverished (sta-tions 5 and 6), unbalanced (stations 3, 4, 7–10, 13, 16, 17and 24), transition to pollution (stations 11, 12, 14, 18,19, 22 and 23) and polluted (stations 15, 20 and 21). Interms of macrobenthic structure, stations from Group Apresented the lowest values of AMBI (Fig. 5) with meanvalues of 0.22 ± 0.23 for Group A2 and 1.66 ± 0.66 forGroup A1. In Group B1 the majority of stations wereclassified as slightly polluted (BC ranging from 2.4 to 3.0)and only station 23 was classified as meanly polluted(BC = 3.5). These values reflected the faunal composition

Fig. 4. Cluster diagram applying the Bray–Curtis similarity inde

of the affinity groups. Stations from group A1 were charac-terised by sensitive, indifferent and tolerant taxa, whilegroup A2 was dominated by sensitive and indifferent taxa(Table 4). Most of dominant taxa from group B1 were tol-erant to pollution, and opportunistic taxa (either first- orsecond-order) increased their abundance in stations fromgroups B2, B3 and B4 (Table 4).

The correlation between abiotic and biotic variablesshowed that only abundance was positively correlated toCorg and Cd, whereas Cd was negatively correlated to J 0

(Table 5). Regarding the marine biotic index AMBI, posi-tive correlations were observed for all the abiotic variablesexcept for phaeo, chl a, Pb and Hg (Table 5).

4. Discussion

According to the environmental parameters, the Obidoslagoon may be divided into two contrasting zones, locatedin the inner and outer areas. The inner zone was character-ized by muddy sediments with low concentrations of metalsand organochlorines when compared to several coastal sys-tems (Laane, 1992) and to fine sediments of urbanised estu-arine systems in Portugal (Vale, 1990; Cortesao and Vale,1995; Castro and Vale, 1995; Monterroso et al., 2003).Besides the low degree of contamination, concentrationsand metal/Al ratios in the muddy sediments from this zoneare relatively uniform. This normalisation is frequentlyused to minimise differences related to grain size distribu-tion (Windom et al., 1989). The uniformity in the upperareas in conjunction with low element/Al ratios suggestedminor metal inputs from localised sources. The PCB con-centrations (one order of magnitude lower than DDTs)corroborate the low industrial influence in the Obidoslagoon. The outer zone of sandy sediments presented verylow concentrations of inorganic and organic contaminants.Strong currents and tidal flooding of the area probablyminimize retention of fine particles and associated contam-inants from upstream areas. The contrast in grain sizebetween upper and lower sediments observed in this study

x after a log (x + 1) transformation of the abundance data.

Table 4Mean abundance (Abd.) and standard deviation (SD) for the ten most abundant taxa in each benthic affinity group

Group A1 Group A2 Group B1

Stations 4, 6, 9 Stations 1, 2, 5 Stations 3, 7, 8, 10, 23, 24

Taxa EG Abd. SD Taxa EG Abd. SD Taxa EG Abd. SD

Hydrobia ulvae III 1.7 3.94 Saccocirrus papillocercus I 4.3 4.07 Hydrobia ulvae III 139.2 142.25Nephtys cirrosa II 1.1 1.27 Scoloplos armiger I 0.6 1.01 Abra ovata III 15.4 21.06Microphthalmus similis II 0.6 0.73 Tellina tenuis I 0.3 0.71 Cerastoderma edule III 13.3 11.14Ophelia neglecta I 0.6 0.88 Lekanesphaera levii NA 0.3 0.71 Notomastus sp. III 12.2 27.59Tellina donacina I 0.4 0.88 Hesionura sp. II 0.2 0.44 Heteromastus filiformis IV 10.9 20.59Timoclea ovata I 0.2 0.44 Hydrobia ulvae III 0.1 0.33 Und. Euclymeninae I 4.3 8.94Lekanesphaera levii NA 0.2 0.67 Nephtys cirrosa II 0.1 0.33 Alkmaria romijni III 4.2 9.79Lepidochitona cinerea NA 0.1 0.33 Protodorvillea kefersteini II 0.1 0.33 Tapes decussatus I 4.1 6.32Hydrobia minuta III 0.1 0.33 Ammodytes tobianus II 0.1 0.33 Pygospio elegans III 3.7 7.02Nassarius reticulatus II 0.1 0.33 Scoloplos armiger I 2.9 4.04

Dominant ecological groups I, II and III Dominant ecological

groups

I and II Dominant ecological

groups

III

Ecological quality status Good to high Ecological quality status High Ecological quality status Mainly good

Group B2 Group B3 Group B4

Stations 15, 21, 22 Stations 16, 17 Stations 11–14, 18–20

Taxa EG Abd. SD Taxa EG Abd. SD Taxa EG Abd. SD

Capitella spp. V 18.3 18.58 Phoronis spp. II 214.2 58.36 Heteromastus filiformis IV 136.7 125.21Notomastus sp. III 13.0 14.52 Corbula gibba IV 147.3 81.19 Und. Oligochaeta V 26.1 26.94Malacoceros fuliginosus V 9.7 12.05 Und. Insecta

(Chironomidae)IV 61.0 93.51 Abra tenuis III 18.9 33.19

Hydrobia ulvae III 8.2 22.08 Heteromastus filiformis IV 52.8 57.72 Phoronis spp. II 10.3 13.05Abra alba III 5.1 6.97 Hydrobia ulvae III 19.7 5.16 Cyathura carinata III 8.4 6.72Und. Anthozoa II 4.0 12.00 Abra alba III 13.0 20.15 Cerastoderma edule III 8.0 13.83Heteromastus filiformis IV 3.4 4.13 Scrobicularia plana III 8.2 12.66 Scrobicularia plana III 6.7 9.09Und. Oligochaeta V 3.4 5.41 Abra ovata III 6.8 8.33 Streblospio sp. III 6.2 9.53Polydora hoplura IV 3.3 3.08 Abra tenuis III 5.7 11.96 Hydrobia ulvae III 2.2 3.44Und. Insecta (Chironomidae) IV 1.9 3.92 Hydrobia minuta III 3.8 7.47 Abra alba III 2.0 3.63

Dominant ecological groups III, V Dominant ecological

groups

II, IV Dominant ecological

groups

II, IV and V

Ecological quality status Moderate Ecological quality status Good Ecological quality status Mainly moderate

Ecological groups (EG) for each species are presented following Grall and Glemarec (1997) and Borja et al. (2000) and correspond to: I. Sensitive topollution; II. Indifferent to pollution; III. Tolerant to pollution; IV. Second-order opportunistic taxa; V. First-order opportunistic taxa (see Borja et al.,2000). NA-ecological group not assigned.


is comparable to that reported by Quintino (1988),although the quantity of marine-derived sand transportedinto the lagoon has presumably changed during the timeinterval between the two studies, presumably causing alter-ations in inter-tidal sand flats of the lower zone. At the timeof this study, the transition between muddy and sandyareas was very pronounced and the mixture of sand andfine particles was confined to a small area seawards thejunction of the inner branches of the lagoon.

In the Obidos lagoon, as the distance from the inletincreased, the first parameter to show considerable differ-ences was Corg. It appears that the longitudinal gradientof organic matter content reflects the preferential accumula-tion of this material in upper areas, as a result of either thedrainage basin and of Ulva sp. decomposition. The increasein organic matter content from down- to upstream areaswas associated with the dominance of species assigned toecological groups IV and V. According to Grall and Glema-rec (1997), these groups include opportunistic species,

which in the case of group V are capable of proliferatingin reduced sediments. Species from these groups can alsotolerate toxic conditions (Grall and Glemarec, 1997). Onthe other hand, species from groups I and II, respectivelysensitive and indifferent species to organic enrichment, aremainly associated with the clean sandy downstream area.From the analysis of the ten most abundant taxa of allthe affinity groups identified, it was possible to observethe dominance of group III. This group comprises speciesthat are tolerant to excess levels of organic matter (Gralland Glemarec, 1997) and their dominance is consideredcharacteristic of estuarine communities from sites withorganic matter inputs (Borja et al., 2000). The applicationof the marine biotic index AMBI to the present data setshowed that the Obidos lagoon presented sites ranging fromunpolluted to meanly polluted. Concerning environmentalparameters it was generally observed that Corg, fineparticles, and phaeopigments tend to increase from down-stream to upstream stations (Group A < Group B1 <

Fig. 5. Variation of the AMBI values for the stations analysed in the Obidos lagoon.


Groups B2–B4). The presence of Saccocirrus papillocercus,Nephtys cirrosa, and Microphthalmus similis, and the lowvalues of both Corg and percentage of fine particles observedin Group A reflected the high hydrodynamic conditions ofthe inlet area, which also presented the highest ecologicalquality status. The intermediate values of Corg observed inthe central area were also associated with intermediate val-ues of the index, reflecting the abundance of some speciesusually characteristic of organically enriched areas, suchas Heteromastus filiformis and Notomastus sp., or tolerantto pollution (e.g., Abra ovata and A. nitida). Sediments fromthe inner area are organically enriched and the most charac-teristic species of this area were H. filiformis, oligochaetes,insects, phoronids and Corbula gibba, which are frequentlyconsidered as opportunistic (Pearson and Rosenberg,1978). In fact, sediments from the inner areas are morereduced and the water presents lower oxygen levels duringthe night (Fonseca et al., 2002). The higher organic carboncontents and the consequent increase of percentage of spe-

Table 5Pearson’s correlation coefficients. Relationship between abiotic and bioticvariables for the Obidos lagoon

Variables S N d J 0 H0 AMBI

Fine particles (%) 0.40 0.55 �0.25 �0.46 �0.10 0.85*

Corg (%) 0.47 0.65* �0.24 �0.50 �0.12 0.80*

Phaeopigments(lg g�1)

0.42 0.45 0.03 �0.22 0.02 0.15

Chlorophyll a(lg g�1)

0.59 0.25 0.32 0.08 0.43 0.12

Mg (%) 0.43 0.46 �0.20 �0.43 �0.06 0.84*

Fe (%) 0.44 0.53 �0.23 �0.47 �0.10 0.82*

Mn (lg/g) 0.51 0.57 �0.14 �0.38 0.04 0.89*

Zn (lg/g) 0.34 0.49 �0.27 �0.35 �0.03 0.77*

Cu (lg/g) 0.42 0.56 �0.25 �0.52 �0.14 0.82*

Cd (lg/g) 0.43 0.71* �0.25 �0.65* �0.28 0.72*

Pb (lg/g) 0.06 0.02 �0.16 0.14 0.22 0.41Ni (lg/g) 0.31 0.45 �0.26 �0.40 �0.09 0.65*

Cr (lg/g) 0.46 0.58 �0.24 �0.53 �0.15 0.83*

Hg (lg/g) 0.49 0.30 0.07 �0.35 �0.04 0.51As (lg/g) 0.45 0.48 �0.19 �0.46 �0.08 0.78*

* Statistically significant at p < 0.05.

cies from groups IV and V were reflected in the increasingvalues of AMBI, classifying this area mainly with meanlypolluted stations. Long et al. (1995) in a work undertakento update the guideline values for several contaminants pro-posed two different values, the lower 10th percentile of theeffects data (Effects Range-Low, ERL) and the median, or50th percentile, of the effects data (Effects Range-Median,ERM). According to those authors, concentrations belowthe ERL value represent minimal effects for the biologicalcommunities, while concentrations between ERL andERM values may occasionally cause effects on the commu-nities and in sites with values higher than ERM the effectsmay be frequently observed. In the present work five con-taminants showed higher concentrations than the ERLvalue: Cu, Cr, As and tDDT presented values that maycause occasional effects in the community’s health; Ni con-centrations were above the ERM, and thus effects are moreprobable. These concentrations were observed in the innerarea of the lagoon, where the ecological quality status waslower. However, as those correspond to total concentra-tions and toxicological studies were not undertaken, bio-availability of those contaminants is unknown. In the caseof some of these being bioavailable for the benthic commu-nity, it is possible that those contaminants may also beshaping macrobenthic communities in this area of thelagoon. However, except for cadmium, which concentra-tions were lower than the ERL value, all the metals werenot correlated to the biotic variables analysed. Only thetotal organic carbon was positively correlated to the abun-dance. On the other hand, the biotic index AMBI showedhigh correlation with the majority of the variables analysed,such as the percentage of fine particles and total organiccarbon. There is a highly significant correlation betweentotal organic carbon in the sediment and AMBI, whichhas already been observed by Muxika et al. (2005). Thepositive correlations observed with the majority of contam-inants may be the result of the preferential association ofmetals with the sediments with higher percentage of finesand organic carbon (Windom et al., 1989). This is sup-ported by the low values observed in the lagoon when com-


pared to other water bodies and by the inexistence of corre-lations between each metal and the biotic variables.

5. Conclusions

The applicability of AMBI to assess the ecological statusof macrobenthic communities from different habitats underseveral types and degrees of disturbance (e.g., organicenrichment, petroleum contamination, heavy metal pollu-tion, sand extraction, hypoxia, fish aquaculture) wasalready tested either in European waters (Muxika et al.,2005) or south-western Atlantic regions (Muniz et al.,2005). The effect of seasonal variability in benthic indiceswas also assessed by Reiss and Kroncke (2005) and thoseauthors found that multimetric indices such as AMBI wereless influenced by the natural seasonal variability of macro-fauna, when compared to univariate indices like Shannon-Wiener index. Nevertheless, the use of different indices isalways advised in order to a better evaluation of thebenthic community health (Muniz et al., 2005; Labruneet al., 2006) and preferentially in association with otherparameters (Borja et al., 2004b,c). In the present workAMBI showed to be suitable for the discrimination of thebiological and environmental gradients observed in thissystem, which is characterised by low concentrations ofcontaminants and high organic matter content in innerareas. Despite the high dominance ratios were observedin all sampling stations, most of the zones analysed didnot present a low quality status. The macrobenthic com-munities’ health generally increased from up- to down-stream stations, where the effect of the organic matter islower. According to the WFD, in a near future all Euro-pean water bodies must be classified as ‘good’ quality sys-tems. From the present data this lagoon presented areasthat not fit within those standards. Regardless the low lev-els of chemical contaminants in all stations, the AMBIindex allowed the identification of three contaminated sta-tions that could not be distinguished by the differences intotal contaminant concentrations.

Acknowledgements

The authors are grateful to Paula Pereira and FranciscoLeitao for their contribution concerning species identifi-cation and to Jorge Santos and Miguel Quintans forassistance provided during the sampling campaigns.Authors are also indebt to the referees for their invaluablecontribution for the improvement of an earlier draftversion.

References

Borja, A., Franco, J., Perez, V., 2000. A marine biotic index to establishthe ecological quality of soft-bottom benthos within Europeanestuarine and coastal environments. Mar. Pollut. Bull. 40, 1100–1114.

Borja, A., Franco, J., Muxika, I., 2004a. The biotic indices and the WaterFramework Directive: the required consensus in the new benthicmonitoring tools. Mar. Pollut. Bull. 48, 405–408.

Borja, A., Franco, J., Valencia, V., Bald, J., Muxika, I., Belzunce, M.J.,Solaun, O., 2004b. Implementation of the European water frameworkdirective from the Basque country (northern Spain): a methodologicalapproach. Mar. Pollut. Bull. 48, 209–218.

Borja, A., Valencia, V., Franco, J., Muxika, I., Bald, J., Belzunce, M.J.,Solaun, O., 2004c. The water framework directive: water alone, or inassociation with sediment and biota, in determining quality standards.Mar. Pollut. Bull. 49, 8–11.

Branco, V., Canario, J., Vale, C., Raimundo, J., Reis, C., 2004. Total andorganic mercury concentrations in muscle tissue of the blue shark(Prionace glauca L. 1758) from the Northeast Atlantic. Mar. Pollut.Bull. 49, 854–874.

Byers, S.C., Mills, E.L., Stewart, L., 1978. A comparison of methods ofdetermining organic carbon in marine sediments, with suggestions fora standard method. Hydrobiologia 58, 43–47.

Carvalho, S., Moura, A., Gaspar, M.B., Pereira, P., Cancela da Fonseca,L., Falcao, M., Drago, T., Leitao, F., Regala, J., 2005. Spatial andinter-annual variability of the macrobenthic communities within acoastal lagoon (Obidos lagoon) and its relationship with environmen-tal parameters. Acta Oecol. 27, 143–159.

Castro, O., Vale, C., 1995. Total PCB-organic matter in sediments fromthree estuarine areas of Portugal. Nether. J. Aquat. Ecol. 29, 297–302.

Chapman, P.M., Dexter, R.N., Long, E.R., 1987. Synoptic measures ofsediment contamination, toxicity and infaunal community composi-tion (the Sediment Quality Triad) in San Francisco Bay. Mar. Ecol.Progr. Ser. 37, 75–96.

Clarke, K.R., Gorley, R.N., 2001. PRIMER v5: User Manual/Tutorial.PRIMER-E. Plymouth, UK, p. 91.

Cortesao, C., Vale, C., 1995. Metals in sediments of the Sado estuary,Portugal. Mar. Pollut. Bull. 30, 34–37.

Fonseca, L.C., Gil, O., Micaelo, C., Pereira, P., Gaspar, M., Antunes, P.,Quintans, M., Falcao, M., Vale, C., Moura, A., Carvalho, S., Leitao,F., Santos, I., Drago, T., Santos, J., Raimundo, J., 2002. Caracter-izacao Ambiental da Lagoa de Obidos, Report (In Portuguese).IPIMAR, Lisbon, Portugal, p. 102.

Gil, O., Vale, C., 2001. Evidence for polychlorinated biphenyls dechlo-rination in sediments of Sado Estuary, Portugal. Mar. Pollut. Bull. 42,453–461.

Grall, J., Glemarec, M., 1997. Using biotic indices to estimate macroben-thic community perturbations in the Bay of Brest. Estuar. Coast. ShelfSci. 44 (Suppl. A), 43–53.

Gray, J.S., 1974. Animal-sediment relationships. Oceanogr. Mar. Biol.:Ann. Rev. 12, 223–261.

Laane, R.W.P.M, 1992. Background concentrations of natural com-pounds in rivers, sea water, atmosphere and mussels. Report DGW-92.033, 8.

Labrune, C., Amouroux, J.M., Sarda, R., Dutrieux, E., Thorin, S.,Rosenberg, R., Gremare, A., 2006. Characterization of the ecologicalquality of the coastal Gulf of Lions (NW Mediterranean). Acomparative approach based on three biotic indices. Mar. Pollut.Bull. 52, 34–47.

Long, E.R., MacDonald, D.D., Smith, S.L., Calder, F.D., 1995. Incidenceof adverse biological effects within ranges of chemical concentrationsin marine and estuarine sediments. Environ. Manage. 19, 81–97.

Moita, I., 1985. Carta dos sedimentos superficiais da Plataforma Conti-nental: Cabo S. Vicente ao Rio Guadiana (SED 7 e 8), 1a edicao.Instituto Hidrografico, Lisboa, Portugal.

Monterroso, P., Pato, P., Pereira, E., Vale, C., Duarte, A.C., 2003.Distribution and accumulation of metals (Cu, Cd, Zn and Pb) insediments of a lagoon on the northwestern coast of Portugal. Mar.Pollut. Bull. 46, 1200–1205.

Muniz, P., Venturini, N., Pires-Vanin, A.M.S., Tommasi, L.R., Borja, A.,2005. Testing the applicability of a Marine Biotic Index (AMBI) toassessing the ecological quality of soft-bottom benthic communities, inthe South America Atlantic region. Mar. Pollut. Bull. 50, 624–637.

Muxika, I., Borja, A., Bonne, W., 2005. The suitability of the marinebiotic index (AMBI) to new impact sources along the European coasts.Ecol. Indicat. 5, 19–31.


Pearson, T.H., Rosenberg, R., 1978. Macrobenthic succession in relationto organic enrichment and pollution of the marine environment.Oceanogr. Mar. Biol.: Ann. Rev. 16, 229–311.

Plante-Cuny, M.R., 1974. Evaluation par spectrophotometrie des teneursen chlorophylle a fonctionnelle et en pheopigments des substratsmeubles marins. Documents Scientifiques de la Mission de l’O.R.S.T.O.M. a Nosy-Be 45, p. 76.

Quintino, V., 1988. Structure et cinetique comparees des communautes demacrofaune benthique de deux systemes lagunaires de la cote ouest duPortugal: (Obidos et Albufeira). Ph.D. thesis. University of Paris 6,Paris, France, p. 333.

Rantala, R.T.T., Loring, D.H., 1977. A rapid determination of 10elements in marine suspended particulated matter by atomic absorp-tion. Atom. Absorpt. Newsletter 16, 51–52.

Rees, H.L., Moore, D.C., Pearson, T.H., Elliott, M., Servile, M., Pomfret,J., Johnson, D., 1990. Procedures for the monitoring of marine benthiccommunities at UK Sewage Sludge Disposal Sites. Dep. Agricultureand Fisheries for Scotland. Scottish Fisher. Informat. Pamphlet 18, 1–79.

Reiss, H., Kroncke, I., 2005. Seasonal variability of benthic indices: anapproach to test the applicability of different indices for ecosystemquality assessment. Mar. Pollut. Bull. 50, 1490–1499.

Rhoads, D.C., Mccall, P.L., Yingst, J.Y., 1978. Disturbance andproduction on the estuarine sea floor. Amer. Scient. 66, 577–578.

Rosenberg, R., Blomqvist, M., Nilsson, H.C., Cederwall, H., Dimming,A., 2004. Marine quality assessment by use of benthic species-abundance distributions: a proposed new protocol within the Euro-pean Union water framework directive. Mar. Pollut. Bull. 49, 728–739.

Rueda, J.L., Fernandez-Casado, M., Salas, C., Gofas, S., 2001. Season-ality in a taxocoenosis of molluscs from soft bottoms in the Bay ofCadiz (southern Spain). J. Mar. Biol. Assoc. UK 81, 903–912.

Simboura, N., Zenetos, A., 2002. Benthic indicators to use in EcologicalQuality classification of the Mediterranean soft bottom marineecosystems, including a new Biotic Index. Mediterranean Mar. Sci.3/2, 77–111.

Snelgrove, P.V.R., Butman, C.A., 1994. Animal-sediment relationshipsrevisited: cause versus effect. Oceanogr. Mar. Biol.: Ann. Rev. 32, 111–117.

Vale, C., 1990. Temporal variations of particulate metals in the TagusRiver Estuary. Sci. Total Environ. 97/98, 137–154.

Windom, H.L., Schropp, S.J., Calder, F.D., Ryan, J.D., Smith, R.G.,Burney, L.C., Lewis, F.G., Rawlinson, C.H., 1989. Natural trace metalconcentration in estuarine and coastal marine sediments of thesoutheastern United States. Environ. Sci. Technol. 23, 314–320.