Impact of sewage pollution on the structure and functioningof a rocky shore benthic community

J. Cabral-OliveiraA,C, M. DolbethA,B and M. A. PardalA

ACentre of Functional Ecology (CFE), Department of Life Sciences,

University of Coimbra, Apartado 3046, 3001-401 Coimbra, Portugal.BCESAM & Biology Department, University of Aveiro, Campus de Santiago,

3810-193 Aveiro PortugalCCorresponding author. Email: joanaco@ci.uc.pt

Abstract. The secondary production of rocky shoremacroinvertebrate assemblages impacted through sewage dischargewas assessed, taking into account the trends of production among dominant species and feeding guilds. The present studywas conducted on the Peniche peninsula (central-western Portuguese coast, temperate region), in three areas: one area near

a sewage discharge and two undisturbed reference areas. Within each area, three intertidal zones were monitored - littoralfringe, eulittoral and sublittoral fringe - by taking seasonal samples during one year. The empirical model of Cusson andBourget (2005) was used to evaluate secondary production. In the littoral fringe, no differences in the production valueswere found between impacted and reference areas. In the eulittoral, sewage discharge seemed to affect the natural

competition between patellidae and barnacles by favouring suspension feeders (barnacles), presumably due to higher foodresources near the sewage. In the sublittoral fringe, near the sewage discharge, an increase in the production values oftolerant species was observed to the detriment of the sensitive species, with higher production levels in the reference areas.

Overall, secondary production was higher in the communities near the sewage affected areas, but this increase was mostlydue to the production of tolerant species. The present study showed that the incorporation of secondary production in thebiological assessment provided further insight into the health of the ecosystem, thus being an important tool for

understanding differences in the functioning of the ecosystem.

Additional keywords: invertebrates, rocky shore, secondary production, sewage.

Received 16 July 2013, accepted 11 November 2013, published online 16 June 2014

Introduction

Coastal areas are exposed to a wide range of anthropogenicimpacts. One of the most widespread and harmful sources of

human disturbance in coastal areas is nutrient enrichment due tosewage discharge (Fraschetti et al. 2006; Halpern et al. 2007). InEurope, sewage can receive secondary (organic matter removed)or tertiary (nutrients and bacteria removed) treatment, and then be

released directly near shore, or at some distance through pipelinesystems (http://epp.eurostat.ec.europa.eu/statistics_explained/index.php/Water_statistics, accessed 20 January 2014).

The impact of sewage pollution on marine habitats has beenstudied by assessing the response of the biological assemblages.One of the ways to measure responses from biological commu-

nities is by comparing species richness. However, there is not aconsistent response in diversity to sewage pollution: in somecases it may increase (Lopez Gappa et al. 1993); decrease(Littler and Murray 1975; Archambault et al. 2001); or reveal

no clear effect (Cabral-Oliveira et al. 2014). Changes in speciesrichness seem to vary according to the type and intensity ofdisturbance, and are related to the original community of the

study area (Whomersley et al. 2010). Other studies have

analysed the effects of sewage pollution on the patterns ofdensity and biomass of biological assemblages. Again, there isno homogeneous response. In some instances, the presence of

sewage discharge led to increased abundance (Lopez Gappaet al. 1993), while in others a decrease was observed (Littler andMurray 1975; Archambault et al. 2001). The most consistentresult was found by observing the effects of sewage on the

composition and structure of the communities (Littler andMurray 1975; Chapman et al. 1995; Fraschetti et al. 2006among others), with the replacement of sensitive species by

tolerant ones. In the light of the current need to includefunctional approaches into the assessment of environmentalimpacts (Elliott and Quintino 2007), the effects of sewage

pollution can also be studied by evaluating changes in theecosystem function (Dolbeth et al. 2012).

Secondary production (production by heterotrophic organ-isms) is the amount of organic matter or energy incorporated in a

given area per time unit; it represents a measure of the function-ing of the ecosystem, as it combines both static and dynamiccomponents of the ecological performance of a population

(Dolbeth et al. 2012). The patterns of secondary production

CSIRO PUBLISHING

Marine and Freshwater Research

http://dx.doi.org/10.1071/MF13190

Journal compilation � CSIRO 2014 www.publish.csiro.au/journals/mfr

are affected by several characteristics of life history such as

body mass, recruitment, age, life span, taxonomy and trophicstatus, and the biomass and density of the population (Cussonand Bourget 2005). As such, measures of secondary productionmay help to assess the fitness of a population; they can addmore

information to diversity and density parameters about changesin the ecosystem and, ultimately, about the food provisiondelivered by an ecosystem (Benke and Huryn 2010; Dolbeth

et al. 2012). The importance of incorporating secondary pro-duction to assess the impacts of sewage pollution and nutrientenrichment has been recognised for freshwater (Whiles and

Wallace 1995; Woodcock and Huryn 2007; Faupel et al. 2012)and estuarine ecosystems (Dolbeth et al. 2011, 2012). However,little is known about the habitats of the rocky shores.

The aim of this work is to understand the effect of sewage

pollution on the secondary production of macroinvertebratefauna of a rocky shore. For that purpose, the secondary produc-tion of benthic macroinvertebrates was studied and compared,

taking into account the trends of production among dominantspecies and feeding guilds, in an impacted area (near a sewagedischarge) and on two reference areas.

Material and methods

Study site and sampling procedures

The present study was conducted in Peniche peninsula, located

in a temperate region on the central western coast of Portugal(Fig. 1). In this peninsula, a sewage treatment plant was built in1998. The outfall releases secondary-treated effluents. It servesa human population of 40 000 and discharges the effluent

directly into the intertidal area of the rocky shore. The lack ofpre-impact data led to the choice of an ACI (after control/impact) experimental design (Chapman et al. 1995; Glasby

1997). Consequently, three sampling areas were selected: oneimpacted area near the sewage discharge (Imp) and two refer-ence areas (R1 and R2) to account for the natural differences

among uncontaminated areas (Fig. 1). All sampling areas had

comparable environmental conditions with regard to slope,orientation, wave exposure and type of substrate. Within eachsampling area, three intertidal zones were sampled: the littoralfringe; the eulittoral; and the sublittoral fringe. This pattern of

zonation has been mentioned for the coastal intertidal areasof the Portuguese (Boaventura et al. 2002). The littoral fringe ischaracterised by the presence of Melarhaphe neritoides and

encrusting lichens; the eulittoral zone is dominated by barnaclesand mussels; and the sublittoral fringe is dominated by red algaein central and southern regions of Portugal (Boaventura et al.

2002).For each intertidal zone, five quadrats (12� 12 cm) were

randomly selected and organismswere collected by scraping theselected area using a spatula and a chisel. As the studied area

presents a temperate climate, four sampling dates representativeof each season were chosen in order to account for the temporalvariation, in February, April, July and November 2010. A total

of 45 samples were collected during each field trip.

Laboratory procedures

All material collected from scrapings was sieved through a

500-mm mesh and all individuals were identified to the lowestpossible taxonomic level and counted to determine density.Biomass was determined as ash-free dry weight (AFDW), after

combustion for 8 h at 4508C.Water samples were filtered (Whatman GF/F glass-fibre

filter) and stored frozen at �188C until analysis. Analysesfollowed the standard methods described in Limnologisk

Metodik (Ferskvandsbiologisk Laboratorium 1985) for ammo-nia (NH4

þ) and phosphate (PO43-), in Strickland and Parsons

(1972) for nitrite (NO2- ), and ESS Method 340.2 for total

suspended solids (Environmental Sciences Section 1993).Trace elements were analysed in the red algae Corallina sp.,

the dominant species in the sublittoral fringe, to search for

SPAIN

PENICHE

ImpCABO CARVOEIRO(36� 21′ N; 9� 24′ W)

0 0.5 1 Km

R2R1

PO

RT

UG

AL

ATLA

NT

IC O

CE

AN

N

Fig. 1. Map of the Peniche peninsula, showing the location of the sampling areas: Imp: impacted

area; R1 and R2: reference areas

B Marine and Freshwater Research J. Cabral-Oliveira et al.

possible contamination by trace metals. Five replicates fromeach area were analysed, using algae with similar sizes in order

to avoid misinterpretation due to size-dependent trends in theaccumulation. Samples with ,1 g of fresh algae were digestedusing 8mL of HNO3 65% and 2mL of H2O2 30%. Algae

were accurately weighed in dry, pre-cleaned Teflon digestionvessels that were sealed and placed in the microwave chamber(Multiwave 3000 – Anton Paar, Austria) for digestion. Analytic

determinations were performed using the Atomic absorptionspectrophotometer (SOLAAR M2, Thermo Unicam, USA).With regards to arsenic (As) contamination specifically, itsconcentration was also determined in the seawater following

the methods described in Bermejo-Barrera et al. (1998), usingfiltered water samples (Whatman GF/F glass-fibre filter,0.45 mm) from all sampling areas.

Data analysis and secondary production estimates

A permutational multivariate analysis of variance (PERMA-

NOVA; Anderson 2001) was carried out separately for eachintertidal zone (littoral fringe, eulittoral and sublittoral fringe); itwas selected to test for the differences in biomass of the inver-tebrate assemblage structure between impacted and reference

conditions, considering each sampling occasion (differentseasons). The model consisted of two factors: Time (four levelscorrespondening to each sampling occasion, random, ortho-

gonal) and Location (one impacted and two reference areas,fixed, orthogonal). In both cases, the design was asymmetricaldue to the presence of a single impacted location (Underwood

1991). Therefore, the Location term, with three levels and twodegrees of freedom, was partitioned into two portions: the testbetween Impacted versus Reference areas (Imp-v-Rs) and

the test of the variability between Reference areas (Rs). TheTime�Location interaction (Ti�Rs) was similarly dividedinto two portions: a Time� Imp-v-Rs and a Time�Rs inter-action. At last, the same partitioning was performed for the

residual variability of observations within Imp (Res Imp) withinRs (Res Rs). Appropriate denominators for F ratios were iden-tified from expected mean squares and tests were constructed

following the logic of asymmetrical design (see Terlizzi et al.2005a). All analyses were based on Bray–Curtis similarity ofsquare-root transformed data, and each term in the analysis

was tested using 4999 random permutations of appropriateunits. For the eulittoral and sublittoral fringe, differences inthe structure of the community among areas were visualisedthrough principal coordinate (PCO) analyses, based on the

Bray–Curtis similarities of the species biomass (annual aver-age). The taxa discriminating reference and impacted areas weredisplayed as vectors in the PCO plots (Spearman correlation;

R . 0.6). For all statistical tests, the significance level was setat P# 0.05

For the assessment of secondary production, an empirical

model was chosen because there was not sufficient data to applyclassicalmethods (i.e. it was not possible to follow the cohorts orestimate the growth for all species (Dolbeth et al. 2012)) due to

the seasonal sampling and because individuals were not mea-sured. However, empirical models may provide fair estimates ofproduction, allowing hypotheses testing, especially because theproduction of the populations was pooled into the production of

the community (Brey 2001; Dolbeth et al. 2012); also, the

sampling procedure employed allows for reasonable estimatesof the annual average biomass and body mass. As such, secon-

dary production was estimated using the Cusson and Bourget(2005) general model, because among empirical models it hasthe highest coefficient of determination to estimate production

in marine systems (R2¼ 0.92, Dolbeth et al. 2012):

log P ¼ 0:45þ 1:01� log �B� 0:84� log L� 0:09� �w

where P is the annual production (kJ.m�2.year�1), �B is theannual mean biomass (kJ.m�2), �w is the annual mean individualmass (kJ.individual�1), and L is the life span (years). Data on �Band �w were obtained by averaging the biomass obtained in thefour sampling occasions, as representative of annual meanbiomass. For the application of the model, information on thespecies life span was collected from the literature (http://www.

marlin.ac.uk, www.nhm.ac.uk, accessed 20 January 2014;http://www.genustraithandbook.org.uk, accessed 20 January2014), and all species were assigned to a functional feeding

group (see Supplementary Material) in order to obtain a dis-crimination of production per feeding guild. For rare speciesor for those with no information on life span, an average value

of life span was inferred from similar species traits (afterCusson and Bourget 2005). Biomass was converted into energy(kJ) using the weight-to-energy ratio provided in Brey (2001).A single value of annual production was obtained for each

sampling area and level.The value of secondary production within the macroinverte-

brate assemblage in the three sampling areas (Imp, R1 and R2)

and intertidal zones (littoral fringe, eulittoral and sublittoralfringe) were explored using a two-way crossed ANOSIM withno replicates. Similarities in the production data were calculated

as the Bray–Curtis coefficient after square-root-transformationof the raw data to scale down the scores of the very productivespecies (Clarke and Warwick 2001). Non-metric Multidi-

mensional Scaling (nm-MDS) was subsequently performed toclarify the patterns, and similarity percentages obtained fromCLUSTER analysis were overlaid in the plot. All the multivari-ate analyses were done using PRIMERv6 and PERMANOVAþroutines (Anderson et al. 2008).

Results

Environmental parameters

The environmental parameters – seawater temperature, dis-solved oxygen, salinity, pH, concentrations of nutrients and totalsuspended solids –weremarkedly different when comparing the

sampling areas (Table 1). The temperature of the seawater washigher in the impacted area, while dissolved oxygen, salinity andpHwere lower in the sewage-affected areas. The concentrations

of nutrients, especially ammonia and phosphate, and total sus-pended solids (TSS)were also higher near the sewage discharge.Regarding trace elements (Table 2), As was the only elementwith different concentration in each one of the two reference

areas and in the impacted area (ANOVA, Dunnett’s test,F(2,12)¼ 106.5, P, 0.05). Moreover, the concentrations of Aswere higher near the sewage discharge (Table 2). Therefore, the

concentration of this metalloid was also measured in the sea-water, with results showing higher concentrations near thesewage discharge (Table 1).

Macroinvertebrate production affected by sewage Marine and Freshwater Research C

Macroinvertebrate assemblage biomass

A total of 115 803 specimens were collected, being distributedamong 70 different invertebrate taxa. In the littoral fringe, only

the gastropod Melarhaphe neritoides and (occasionally) theisopod Ligia oceanica were found. The eulittoral zone wasdominated by barnacles (Chthamalus montagui), limpets andsmall littorinids. The sublittoral fringe was dominated by the red

algae Corallina spp. that serves as the habitat for a variety ofgastropods, bivalves, crustaceans, polychaetes and chitons.

PERMANOVA analysis detected significant differences

between impacted and reference areas (Imp-v-Rs) in the eulit-toral and sublittoral fringe (Table 3). Moreover, results showedthat such patterns were not consistent in time as a significant

Ti� Imp-v-Rs interaction was detected (Table 3). With regardto the littoral fringe, there were no significant differencesbetween impacted and reference areas (Table 3). Consistent

with the results of the PERMANOVA, the PCO plot showed aclear separation between impacted and reference area samplesfor both the sublittoral fringe and eulittoral zones. In thesublittoral fringe, Patella spp., Mytilus galloprovincialis and

Lasae adansoni were related to the impacted area, while Rissoa

parva, Runcina coronata and Pirimela denticulata discrimi-nated both reference areas (Fig. 2a). In the eulittoral zone, there

was a better differentiation between disturbed and referencesamples, with higher biomass of Chthamalus montagui,Melar-

haphe neritoides, orM. galloprovincialis found in the impactedarea (Fig. 2b).

Secondary production estimates

The patterns of secondary production were studied for eachintertidal zone at each of the three sampling areas (Fig. 3). The

highest values of production were observed in the impactedareas of the sublittoral fringe and eulittoral, whereas the lowestvalues were consistently observed in the littoral fringe, inde-

pendently from impact (Fig. 3a). However, different speciescontributed to the overall production of the community amongthe tidal levels. In the sublittoral fringe, fewer species contrib-

uted to the production levels near the sewage discharge(Fig. 3b). In the reference areas, the dominant contributors to thetotal secondary production were Rissoa parva, Modiolus

modiolus and Mytilus galloprovincialis, whereas in the

impacted area ,50% of the community production was due to

Table 1. Variation of environmental parameters in the three sampling areas

Data are themean� s.d. R1, R2: reference areas; Imp: impacted area; Temp: seawater temperature; TSS: total suspended solids; O2: dissolved oxygen; salinity;

pH and nutrients: NH4: ammonia; PO4: phosphate; Si: silica; NO3: nitrate; NO2: nitrite

Temp (8C) O2 (mg/L) Salinity pH Nutrients TSS/L [As](mg/L)

NH4þ PO4

3� Si NO2�

R1 15.4� 1.8 9.4� 0.5 35.9� 0.3 8.0� 0.2 0.07� 0.03 0.02� 0.01 0.6� 0.3 0.02� 0.01 24.8� 12.5 0.83

R2 15.9� 1.8 9.4� 0.8 35.9� 0.3 8.1� 0.1 0.04� 0.03 0.02� 0.01 0.5� 0.4 0.02� 0.01 13.2� 7.4 1.16

Imp 18.4� 1.7 6.9� 3.5 28.0� 7.1 7.9� 0.3 1.1� 1.0 2.5� 2.3 1.6� 1.0 0.10� 0.01 61.3� 26.4 5.54

Table 2. Concentration of trace metals (lg.g21dry wt ± s.d.) in soft tissues of Corallina sp

Ni Cu Cd Co Pb Zn Fe Mn As

R1 13.5� 0.5 6.9� 1.0 0.6� 0.1 12.9� 0.8 18.4� 1.0 40.6� 4.5 578.7� 123.0 48.0� 5.5 2.2� 0.7

R2 12.7� 0.9 7.4� 0.5 0.6� 0.1 12.3� 0.8 18.6� 1.7 30.3� 3.2 1014.8� 314.0 32.5� 4.2 2.8� 0.7

Imp 12.8� 0.5 4.7� 0.6 0.7� 0.1 12.9� 0.2 16.8� 0.1 34.9� 3.3 406.7� 65.6 23.9� 1.7 28.2� 5.5

Table 3. PERMANOVA results on biomass of macrofauna assemblages for the different intertidal levels (littoral fringe,

eulittoral and sublittoral fringe)

Significant results are given in bold (see text for further details)

Source of variability d.f. Littoral fringe Eulittoral Sublittoral fringe

MS F P MS F P MS F P

Time¼Ti 3 2.0261 3.4277 0.0202 1331.30 3.7826 0.0004 6682.3 5.9970 0.0002

Location¼Lo 2 0.9126 1.7427 0.2612 9506.0 12.347 0.0058 11236 3.6895 0.0078

Imp-v-Rs 1 0.1331 0.1320 0.8150 18083.0 13.228 0.0268 19815 5.9757 0.0276

Rs 1 1.6921 43.377 0.0322 929.06 5.3781 0.0550 2656.9 0.9575 0.4322

Ti�Lo 6 0.5237 0.8859 0.5132 769.87 2.1874 0.0036 3045.4 2.7330 0.0002

Ti� Imp-v-Rs 3 1.0083 1.7372 0.1740 1367.0 3.8756 0.0002* 3315.9 2.6747 0.0002*

Ti�Rs 3 0.0390 0.0599 0.9844 172.75 0.7153 0.6656 6800.7 4.7932 0.0002

Res 48 0.5911 351.96 1114.3

Res Imp 16 0.4722 572.89 505.2

Res Rs 32 0.6506 241.49 1418.8

D Marine and Freshwater Research J. Cabral-Oliveira et al.

M. galloprovincialis alone (Fig. 3b). In the eulittoral area, anopposite pattern was found with fewer species contributing to

the overall production of the community in the referenceareas. Moreover, a change in the dominant species wasobserved, with Patella spp. found dominant in the reference

areas andChthamalusmontagui in the impacted one. The littoralfringe was dominated by Melarhaphe neritoides in all threeareas. These differences in the composition of species andrespective contributions to the total production of the com-

munity were statistically significant with regard to tidal levels(ANOSIM, R¼ 1, P, 0.05), but not between impacted andreference areas (ANOSIM, R¼ –0.03, P. 0.05). However, a

separation in the nm-MDS was visible when comparing refer-ence and impacted areas in the eulittoral and sublittoral fringes(only 40% similarity, Fig. 4).

With regard to the feeding guilds, differences were alsoobserved between reference and impacted areas. Near thesewage discharge, the feeding guild that most contributed to

total production was the suspension feeders, while in the refer-ence areas the herbivores were the group that most contributedto production levels (Fig. 3c). When comparing tidal levels,feeding guild composition for the sublittoral and eulittoral was

similar, dominated by herbivores and suspension feeders andwith a smaller percentage of omnivores, especially in thesublittoral (Fig. 3c). A single feeding guild was found in

the littoral fringe: the herbivore Melarhaphe neritoides.

Discussion

The sewage treatment plant in Peniche Peninsula is preparedonly for secondary treatment and as a consequence the envi-ronmental parameters changed considerably due to the sewagedischarge. Those differences were predictable (e.g. Lopez

Gappa et al. 1993; Roberts et al. 1998), and resulted in higher

temperature of the seawater, higher concentrations of nutrientsand suspended solids, and lower dissolved oxygen, salinity and

pH in the sewage-affected areas. The sewage treatment plantalso receives industrial effluents, which may explain the highlevels of arsenic (As) contamination.

The impact of sewage discharge was noticed not only in theenvironmental parameters, but also in the biological communi-ties. An increase in the biomass of benthic macrofauna wasobserved in the impacted area, together with changes in the

composition and structure of the communities, especially forthe eulittoral and sublittoral fringe. In both tidal levels, tolerantspecies, such as Mytilus galloprovincialis or Chthamalus mon-

tagui were dominant in the impacted area. In contrast, in thesublittoral fringe, several sensitive species, such asRissoa parvaor Runcina coronata, were clearly associated with the reference

areas. These results are in accordance with several studiesdealing with sewage pollution in rocky shore environments(Littler and Murray 1975; Terlizzi et al. 2005b; Atalah and

Crowe 2012), where a predominance of tolerant species inimpacted areas was observed. However, few have quantifiedthe effects of the sewage discharge on the functioning ofecosystems, and this leads us to the question: is sewage pollution

affecting the production of the macroinvertebrate assemblagesof rocky shores?

The secondary production of the macroinvertebrate commu-

nity was higher near the sewage discharge for the sublittoral andeulittoral zones, but no differences were found in the littoralfringe, most probably because the littoral fringe is located

further away from the sewage discharge. Differences in thecomposition of species and their respective contribution tothe total production of the community were statistically signifi-cant only between tidal levels, but there was a discrimination

between reference and impacted areas (.40% similarity). The

PCO1 (29.4% of total variation)

�80

�60

�40

�20

0

20

40

60

PC

O2

(15.

6% o

f tot

al v

aria

tion)

�60 �40 �20 0 20 40 60

PCO1 (53.6% of total variation)

�80

�60

�40

�20

0

20

40

60

PC

O2

(24.

5% o

f tot

al v

aria

tion)

ImpR1R2

Melarhaphe neritoides

Lasae adansoni

Chthamalus montagui

(b)(a)

�60 �40 �20 0 20 40 60

Patella spp

Lasae adansoni

Modiolus modiolusMytilus galloprovincialis

Mytilus galloprovincialis

Campecopea hirsuta

Rissoa parva

Runcina coronataPirimela denticulata

Fig. 2. Principal Coordinates Ordination (PCO) plots at both impacted (filled symbols) and reference areas (empty symbols) on the basis of Bray–Curtis

similarities of the square-root transformed data: (a) sublittoral fringe assemblages; and (b) eulittoral assemblages.

Macroinvertebrate production affected by sewage Marine and Freshwater Research E

mentioned increase in the secondary production of the commu-nity in the impacted area was probably related to nutrient

enrichment from the sewage discharge. Increases in productionassociated with nutrient enrichment have been observedbefore; intermediate levels of nutrient enrichment in relativelyimpoverished systems can increase primary production and

subsequently secondary production (Nixon and Buckley 2002;Singer and Battin 2007). However, it is important to identify the

source of the production, in particular the species responsible forthe increase, especially because an increase in secondary pro-

duction does not necessarily represent a healthier ecosystem(Dolbeth et al. 2012). In the sublittoral fringe and eulittoral,there were differences in the species that largely contributed tothe total production of the community. Near the sewage dis-

charge, Patella spp. was replaced with Chthamalus montagui inthe eulittoral, while in the sublittoral fringe production was

%

100OthLarAca Aca

LarOth Oth

80

60

40

20

0

Total annual production

Lit

Species contribution (%) for production

Ch

– Reference areas

– Impacted area

Pat – Patella sp

Mel – Melarhaphe neritoides

%Feeding guilds contribution (%) for production

Sublittoral fringe

Suspfeed

Eulittoral Littoral fringe

R1

Legend:Herb – Herbivores Susp feed – Suspension feeders Omn – Omnivores

Imp

MytMytMyt

Mod

ModModRis

PatPat

Pat Pat

PatPat

Ris

1.2

1.0

0.8

0.6

0.4

0.2

0

100

80

60

40

20

0R2 R1 ImpR2 R1 ImpR2

R1 ImpR2 R1 ImpR2 R1 ImpR2

Sublittoral fringe Eulittoral Littoral fringe

Suspfeed

Omn OmnOmn

Suspfeed

Suspfeed

Suspfeed

Suspfeed

Aca – Acanthochitona sp

Ch – Chathamalus montagui Myt – Mytilus galloprovincialis Ris – Rissoa parva

Lar – Insecta (larvae) Mod – Modiolus modiolus Oth – Other species

Legend:

Subl Eul

Ch Myt

ChMel Mel Mel

Herb Herb Herb Herb Herb Herb

Herb Herb Herb

P (

gAF

DW

.m�

2 .y

r�1 )

(a)

(b)

(c)

Fig. 3 (a) Total community annual production at the sublittoral fringe (Subl), eulittoral (Eul) and littoral fringe (Lit) in the three sampling areas: reference

areas (empty bars) andimpacted area (filled bars). (b) Species contribution (%) for production and ,feeding guilds contribution (%) to production in each

level in the three sampling areas. R1, R2:reference areas; Imp:Impacted area.

F Marine and Freshwater Research J. Cabral-Oliveira et al.

dominated by the bivalve M. galloprovincialis, revealing a

considerable decrease in species such as Rissoa parva andAcanthochitona spp. These dominant species in the impactedarea, C. montagui and M. galloprovincialis, are tolerant to

nutrient enrichment and trace elements (Borja et al. 2000; http://www.marlin.ac.uk/speciessensitivity.php?speciesID=2981 andhttp://www.marlin.ac.uk/speciessensitivity.php?speciesID=3848, accessed in 20 January 2014), while R. parva and

Acanthochitona spp. are sensitive (Terlizzi et al. 2005b;Atalah and Crowe 2012) and Patella spp. is moderately sensi-tive to decreases in salinity (http://www.marlin.ac.uk). In fact,

M. galloprovincialis has been associated to moderate-poorenvironmental quality status and Patella spp. to high-goodstatus in assemblages of rocky shores (Dıez et al. 2012). In

addition,M. galloprovincialis may have an important commer-cial value (Food and Agriculture Organisation of the UnitedNations (FAO 2012). Therefore, an increase in the secondaryproduction of this species could represent a benefit from an

economic perspective. However, this increased production hasbeen generated from polluted and potentially As-contaminatedareas, thus it is inappropriate for human consumption. So,

although there was a general increase in the secondary produc-tion of the community in the sewage-affected areas, this increasewas due to the tolerant species to the detriment of the sensitive

ones, a pattern that has already been noticed for other habitatsunder stress due to nutrient enrichment (Dolbeth et al. 2003;Singer and Battin 2007) and contaminants (Whiles andWallace

1995; Woodcock and Huryn 2007).The feeding guild composition also changed in response to

sewage impact. Boaventura et al. (1999) studied the trophicstructure of macrobenthic assemblages from rocky shores

along the Portuguese coast. These authors found that filterfeeders were numerically dominant at shallower depths, beingreplaced by detritivores at the deepest areas, whereas both

herbivores and carnivores became abundant in the intermediateareas (Boaventura et al. 1999). In the present study, wherefeeding guilds were quantified by production and not by

density, the eulittoral and sublittoral fringe were dominated

by herbivores, followed by suspension (or filter) feeders in thereference areas. In the impacted area, suspension feeders were

the main contributor for the total production of the community.This dominance of suspension feeders may be related to thehigher amount of suspended solids near the sewage discharge,

and thus higher availability of food for suspension feeders.Previous studies pointed out that suspension feeders are anexample of optimal foraging in the marine context (Gili and

Coma 1998), and that under favourable conditions (nutrientenrichment and tolerance to the changes in the environmentalparameters) species belonging to this feeding guild can achieveextremely high abundance. For the reference areas, in the

eulittoral, the feeding guild that most contributed to totalproduction was the herbivores, while suspension feeders suchas the barnacle C. montagui decreased considerably, probably

due to reduced food resources and lower temperature (Riley2002). This decrease in the secondary production of barnaclesmight diminish competition for space, enabling herbivore spe-

cies such as Patella spp. to attain higher production levels.To sum up, the presence of sewage discharge seems to affect

the eulittoral and sublittoral levels negatively. In the eulittoral, itappeared to affect the natural competition between patellidae

and barnacles, by clearly favouring the barnacles. In the sublit-toral fringe, sewage increased the production of tolerant speciesand reduced the presence of sensitive ones (such as gastropods

or chitons).Overall, the present study showed that the incorporation of

secondary production in the biological assessment of sewage-

impacted areas provided further insight into the health of theecosystem, thus being an important tool for understandingdifferences in the functioning of the ecosystem and, ultimately,

to propose measures to prevent further deterioration of marinehabitats. Secondary production is directly related to consump-tion of food resources (Benke and Huryn 2010), but theincreased production achieved in the impacted areas was essen-

tially due to tolerant and As-contaminated species, which mightsomewhat compromise the integrity of the associated rocky-shore food web. The assessment of specific-species production

(Dolbeth et al. 2012) helped to perceive the former pattern and tounderstand the role of tolerant and sensitive species better.

Acknowledgements

We wish to thank all the colleagues that helped in the field and labora-

tory work. This work was supported by FCT (Fundacao para a Ciencia

e Tecnologia) through a PhD grant attributed to J. Cabral-Oliveira (SFRH/

BD/48874/2008), with funds from POPH (Portuguese Operational Human

Potential Program), QREN Portugal (Portuguese National Strategic Refer-

ence Framework) and MCTES (Portuguese Ministry of Science, Tech-

nology, and Higher Education).

References

Anderson, M. J. (2001). A new method for non-parametric multivariate

analyses of variance. Austral Ecology 26, 32–46.

Anderson, M. J., Ray, N. G., and Clarke, K. R. (2008). PERMANOVA

for PRIMER: guide to software and statistical methods. (PRIMER–E :

Plymouth.)

Archambault, P., Banwell, K., and Underwood, A. J. (2001). Temporal

variation in the structure of intertidal assemblages following the removal

of sewage. Marine Ecology Progress Series 222, 51–62. doi:10.3354/

MEPS222051

Sublittoral fringeEulittoralLittoral fringe

Similarity60%40%

R1R2

Imp

R1R2

Imp

R1 R2

Imp

2D Stress: 0

Legend:

Fig. 4. Two dimensional nm-MDS ordination plot of macroinvertebrate

assemblage production for the reference (R1 and R2) and impacted (Imp)

areas in the sublittoral fringe, eulittoral and littoral fringe, with indication of

the 40% and 60% similarity groups.

Macroinvertebrate production affected by sewage Marine and Freshwater Research G

Atalah, J., and Crowe, T. P. (2012). Nutrient enrichment and variation in

community structure on rocky shores: The potential of molluscan

assemblages for biomonitoring. Estuarine, Coastal and Shelf Science

99, 162–170. doi:10.1016/J.ECSS.2011.12.034

Benke, A. C., and Huryn, A. D. (2010). Benthic invertebrate production:

facilitating answers to ecological riddles in freshwater ecosystems.

Journal of the North American Benthological Society 29, 264–285.

doi:10.1899/08-075.1

Bermejo-Barrera, P.,Moreda-Pifieiro, J., Moreda-Pifieiro, A., and Bermejo-

Barrera, A. (1998). Direct determination of Arsenic in sea water by

electrothermal atomization atomic absorption spectrometry using D2

and Zeeman background correction. Mikrochimica Acta 128, 215–221.

doi:10.1007/BF01243052

Boaventura, D., Fonseca, L. C., and Teles-Ferreira, C. (1999). Trophic

structure of macrobenthic communities on the Portuguese coast.

A review of lagoonal, estuarine and rocky littoral habitats. Acta Oeco-

logica 20, 407–415. doi:10.1016/S1146-609X(99)00127-7

Boaventura, D., Re, P., Cancela da Fonseca, L., and Hawkins, S. J. (2002).

Intertidal rocky shore communities of the continental Portuguese coast.

Comparative analysis of distribution patterns. Marine Ecology (Berlin)

23, 69–90. doi:10.1046/J.1439-0485.2002.02758.X

Borja, A., Franco, J., and Perez, V. (2000). A marine biotic index to the

establish ecology quality of soft-bottom benthos within European

estuarine coastal environments. Marine Pollution Bulletin 40,

1100–1114. doi:10.1016/S0025-326X(00)00061-8

Brey, T. (2001). Population Dynamics in Benthic Invertebrates. A Virtual

Handbook.Version 01.2. (AlfredWegener Institute for Polar andMarine

Research: Germany.) Available at http://www.thomas-brey.de/science/

virtualhandbook/intro/tech.html [Verified 20 January 2014]

Cabral-Oliveira, J., Mendes, S., Maranhao, P., and Pardal, M. A. (2014).

Effects of sewage pollution on the structure of rocky shore macroinver-

tebrate assemblages. Hydrobiologia 726, 271–283.

Chapman, M. G., Underwood, A. J., and Skilleter, G. A. (1995). Variability

at different spatial scales between a subtidal assemblage exposed to the

discharge of sewage and two control assemblages. Journal of Experi-

mental Marine Biology and Ecology 189, 103–122. doi:10.1016/0022-

0981(95)00017-L

Clarke, K. R., andWarwick, R. M. (2001). Change in Marine Communities.

An Approach to Statistical Analysis and Interpretation. 2nd Edn.

(Primer-E: Plymouth.)

Cusson, M., and Bourget, E. (2005). Global patterns of macroinvertebrate

production in marine benthic habitats.Marine Ecology Progress Series

297, 1–14. doi:10.3354/MEPS297001

Dıez, I., Bustamante, M., Santolaria, A., Tajadura, J., Muguerza, N., Borja,

A., Muxika, I., Saiz-Salinas, J. I., and Gorostiaga, J. M. (2012).

Development of a tool for assessing the ecological quality status of

intertidal coastal rocky assemblages, within Atlantic Iberian coasts.

Ecological Indicators 12, 58–71. doi:10.1016/J.ECOLIND.2011.05.014

Dolbeth, M., Pardal, M. A., Lillebø, A. I., Azeiteiro, U. M., and Marques,

J. C. (2003). Short- and long-term effects of eutrophication on the

secondary production of an intertidal macrobenthic community.Marine

Biology 143, 1229–1238. doi:10.1007/S00227-003-1133-5

Dolbeth, M., Cardoso, P. G., Grilo, T. F., Bordalo, M. D., Raffaelli, D., and

Pardal, M. A. (2011). Long-term changes in the production of estuarine

macrobenthos affected by multiple stressors. Estuarine, Coastal and

Shelf Science 92, 10–18. doi:10.1016/J.ECSS.2010.12.006

Dolbeth, M., Cusson, M., Sousa, R., and Pardal, M. A. (2012). Secondary

production as a tool for better understanding of aquatic ecosystems.

Canadian Bulletin of Fisheries and Aquatic Sciences 69, 1230–1253.

doi:10.1139/F2012-050

Elliott, M., and Quintino, V. (2007). The estuarine quality paradox, environ-

mental homeostasis and the difficulty of detecting anthropogenic stress

in naturally stressed areas. Marine Pollution Bulletin 54, 640–645.

doi:10.1016/J.MARPOLBUL.2007.02.003

Environmental Sciences Section (ESS), Inorganic Chemistry Unit,

Wisconsin State Lab of Hygiene (1993). ESS Method 340.2: Total

Suspended Solids, Mass Balance (Dried at 103–105 8C) Volatile

Suspended Solids (Ignited at 550 8C). Wisconsin State Lab of Hygiene,

Madison, WI.

Food and Agriculture Organization of the United Nations (FAO). (2012).

Fisheries and Aquaculture Department. Available from: http://

www.fao.org/fishery/culturedspecies/Mytilus_galloprovincialis/en

[Access 12 November 2012]

Faupel, M., Ristau, K., and Traunspurger, W. (2012). The functional

response of a freshwater benthic community to cadmium pollution.

Environmental Pollution 162, 104–109. doi:10.1016/J.ENVPOL.2011.

11.004

Ferskvandsbiologisk Laboratorium (1985). Limnologisk Metodik. Koben-

hawns Universitet (ed.), (Akademisk Forlag: Copenhagen.)

Fraschetti, S., Gambi, C., Giangrande, A., Musco, L., Terlizzi, A., and

Danovaro, R. (2006). Structural and functional response of meiofauna

rocky assemblages to sewage pollution. Marine Pollution Bulletin 52,

540–548. doi:10.1016/J.MARPOLBUL.2005.10.001

Gili, J. M., and Coma, R. (1998). Benthic suspension feeders: their para-

mount role in littoral marine food webs. Trends in Ecology & Evolution

13, 316–321. doi:10.1016/S0169-5347(98)01365-2

Glasby, T. M. (1997). Analysing data from post-impact studies using

asymmetrical analyses of variance: a case study of epibiota on marinas.

Australian Journal of Ecology 22, 448–459. doi:10.1111/J.1442-9993.

1997.TB00696.X

Halpern, B. S., Selkoe, K. A., Micheli, F., and Kappel, C. V. (2007).

Evaluating and Ranking the Vulnerability of Global Marine Ecosystems

to Anthropogenic Threats. Conservation Biology 21, 1301–1315.

doi:10.1111/J.1523-1739.2007.00752.X

Littler, M. M., and Murray, S. N. (1975). Impact of sewage on the

distribution, abundance and community structure of rocky inter-

tidal macro-organisms. Marine Biology 30, 277–291. doi:10.1007/

BF00390633

Lopez Gappa, J. J., Tablado, A., and Magaldi, N. H. (1993). Seasonal

changes in an intertidal community affected by sewage pollution.

Environmental Pollution 82, 157–165. doi:10.1016/0269-7491(93)

90113-3

Nixon, S.W., and Buckley, B. A. (2002). ‘‘A strikingly rich zone’’—nutrient

enrichment and secondary production in coastal marine ecosystems.

Estuaries 25, 782–796. doi:10.1007/BF02804905

Riley, K. (2002). Chthamalus montagui. Montagu’s stellate barnacle.

Marine Life Information Network: Biology and Sensitivity Key

Information Sub-programme. Marine Biological Association of the

United Kingdom, Plymouth. Available from: http://www.marlin.ac.uk/

speciesbenchmarks.php?speciesID=2981 [Accessed 12 November

2012]

Roberts, D. E., Smith, A., Ajani, P., and Davis, A. R. (1998). Rapid changes

in encrustingmarine assemblages exposed to anthropogenic point source

pollution: a ‘Beyond BACI’ approach.Marine Ecology Progress Series

163, 213–224. doi:10.3354/MEPS163213

Singer, G. A., and Battin, T. J. (2007). Anthropogenic subsidies alter

stream consumer–resource stoichiometry, biodiversity, and food chains.

Ecological Applications 17, 376–389. doi:10.1890/06-0229

Strickland, J. D. H., and Parsons, T. R. (1972). ‘A Practical Handbook of

Seawater Analysis, 2nd edn.’ Fisheries Research Board of Canada

Bulletin 167, (The Alger Press: Ottawa, Canada.)

Terlizzi, A., Benedetti-Cecchi, L., Bevilacqua, S., Fraschetti, S., Guidetti, P.,

and Anderson, M. J. (2005a). Multivariate and univariate asymmetrical

analyses in environmental impact assessment: a case study of Mediter-

ranean subtidal hard substrate sessile assemblages. Marine Ecology

Progress Series 289, 27–42. doi:10.3354/MEPS289027

Terlizzi, A., Scuderi, D., Fraschetti, S., and Anderson, M. J. (2005b).

Quantifying effects of pollution on biodiversity: a case study of highly

H Marine and Freshwater Research J. Cabral-Oliveira et al.

diverse molluscan assemblages in the Mediterranean. Marine Biology

148, 293–305. doi:10.1007/S00227-005-0080-8

Underwood, A. J. (1991). Beyond BACI: experimental designs for detecting

human environmental impacts on temporal variations in natural popula-

tions. Australian Journal of Marine and Freshwater Research 42,

569–587. doi:10.1071/MF9910569

Whiles, M. R., andWallace, J. B. (1995). Macroinvertebrate production in a

headwater stream during recovery from anthropogenic disturbance and

hydrologic extremes. Canadian Bulletin of Fisheries and Aquatic

Sciences 52, 2402–2422. doi:10.1139/F95-832

Whomersley, P., Huxham, M., Bolam, S. G., Schratzberger, M., Augley, J.,

and Ridland, D. (2010). Response of intertidal macrofauna to multiple

disturbance types and intensities—an experimental approach. Marine

Environmental Research 69, 297–308. doi:10.1016/J.MARENVRES.

2009.12.001

Woodcock, T. S., and Huryn, A. D. (2007). The response of macro-

invertebrate production to a pollution gradient in a headwater stream.

Freshwater Biology 52, 177–196. doi:10.1111/J.1365-2427.2006.

01676.X

www.publish.csiro.au/journals/mfr

Macroinvertebrate production affected by sewage Marine and Freshwater Research I