Impacts of the invasive alga Sargassum muticum on ecosystem functioning and food web structure

ORIGINAL PAPER

Impacts of the invasive alga Sargassum muticumon ecosystem functioning and food web structure

Tania Salvaterra • Dannielle S. Green •

Tasman P. Crowe • Eoin J. O’Gorman

Received: 8 February 2012 / Accepted: 11 April 2013 / Published online: 17 April 2013

� Springer Science+Business Media Dordrecht 2013

Abstract Biological invasions have the potential to

cause severe alterations to the biodiversity of natural

ecosystems. At the same time, variation in the diversity

and composition of native communities may have an

important influence on the impact of invasions. Here,

effects of the invasive Japanese wireweed, Sargassum

muticum, were tested across a range of native marine

algal assemblages using a combined additive and

substitutive design. The invasive alga significantly

reduced primary production, an important component

of ecosystem functioning, and increased connectance,

a key property of the food webs associated with the

algal resources. These impacts were mediated by

changes in the proportions of intermediate and top

species, as well as apparent reductions in faunal species

richness and diversity. Some key alterations to faunal

species composition (including the arrival of generalist

species associated with S. muticum) may have been

important in determining these patterns. Overall results

suggest that S. muticum not only directly impeded the

native algal community, but that these effects extended

indirectly to the native fauna and therefore caused

major changes throughout the ecosystem.

Keywords Invasive species � Bottom-up control �Macroalgal communities � Generalist � Robustness �Stability � Ecosystem process rates � Ecological

networks

Introduction

Invasive species may have strong effects on native

communities through processes such as competition,

Electronic supplementary material The online version ofthis article (doi:10.1007/s10530-013-0473-4) containssupplementary material, which is available to authorized users.

T. Salvaterra � D. S. Green � T. P. Crowe �E. J. O’Gorman (&)

School of Biology and Environmental Science,

Science Centre West, University College Dublin,

Belfield, Dublin 4, Ireland

e-mail: [email protected]

T. Salvaterra

e-mail: [email protected]

D. S. Green

e-mail: [email protected]

T. P. Crowe

e-mail: [email protected]

T. Salvaterra

Department of Biology, University of Aveiro, Campus

Universitario de Santiago, 3810-193 Aveiro, Portugal

E. J. O’Gorman

School of Biological and Chemical Sciences, Queen Mary

University of London, Mile End Road, London E1 4NS,

UK

E. J. O’Gorman

Imperial College London, Silwood Park Campus,

Buckhurst Road, Ascot, Berkshire SL5 7PY, UK

123

Biol Invasions (2013) 15:2563–2576

DOI 10.1007/s10530-013-0473-4

parasitism, hybridisation or habitat alteration (Levin

et al. 2002), with positive as well as negative effects

reported (de Wit et al. 2001; Colautti and MacIsaac

2004). Although negative impacts of invaders on

species diversity and abundance are well documented

(Crooks 2002; Grosholz 2002; Levin et al. 2002;

Gribben and Wright 2006; Thomsen et al. 2009), their

effects on ecosystem processes are comparatively

poorly understood (Schaffelke and Hewitt 2007).

Impacts on food web structure and system-level

properties are also likely (Valentine et al. 2002), but

studies highlighting the potential for these effects to

occur are scarce (Grosholz 2002; Baiser et al. 2010;

Carey and Wahl 2010).

The impacts of invasive species depend not only on

the biological traits of the invader, but also on the

characteristics of the recipient communities (Cacabe-

los et al. 2010). By regulating the competitive

environment the invader faces in terms of resources,

species composition and density (Kennedy et al.

2002), native assemblages may have an important

role to play in mitigating the impacts of invaders.

Moreover, when an invasive species becomes estab-

lished in a native community, it may either add to the

overall density of organisms in that community (i.e.

individual plants and/or animals) or replace some of

the native individuals such that the overall density

remains approximately unchanged. Variation in the

extent to which invaders replace native species can

therefore determine the severity of their impacts

(Stachowicz et al. 2002; Fridley et al. 2007). Tests

of the influence of such variation are rare, but can be

made by comparing additive and substitutive (or

replacement series) treatments in experimental frame-

works developed for biodiversity-ecosystem function-

ing research (O’Connor and Crowe 2005; Griffen

2006; Carey and Wahl 2010).

Marine systems are particularly vulnerable to bio-

logical invasions due to their open nature, the naturally

wide geographical ranges involved, and the great

dispersal potential of many marine species (Rapoport

1994). With transoceanic shipping, recreational ves-

sels and importations for aquaculture, the transfer and

dispersal of non-indigenous species outside their

native habitat is greatly enhanced. Once established,

they may bring about large changes in native commu-

nities (Grosholz 2002). As such, marine invaders are

considered a major component of global change,

particularly in coastal systems (Fridley et al. 2004).

Seaweeds are one of the most prominent alien

marine taxa in coastal systems worldwide (Sanchez

and Fernandez 2005; Schaffelke and Hewitt 2007;

Irigoyen et al. 2011). When introduced into the

recipient ecosystem, they tend to have negative

impacts by altering ecosystem processes, developing

into ecosystem engineers and altering food webs

(Schaffelke and Hewitt 2007). Additionally, seaweeds

can disperse efficiently beyond their initial points of

introduction, rapidly becoming abundant, monopolis-

ing space and resources, and reducing the diversity and

biomass of the native algae (Schaffelke and Hewitt

2007). Effects on the native benthic fauna can also

occur, as seaweeds provide habitat and food for these

organisms, determining their patterns of distribution,

abundance and size structure (Cacabelos et al. 2010).

Ultimately, this kind of disturbance to ecosystems and

their constituent species may cause changes to the

native biodiversity and community structure (Schei-

bling and Gagnon 2006), with strong potential for

effects on ecosystem structure and functioning, lead-

ing to degradation and loss of habitat. Consequently, it

is important to understand the potential system-level

impacts of invasive algae in the marine realm.

Of particular concern is the alga Sargassum mut-

icum (or Japanese wireweed), which has become

increasingly widespread and is considered to be one of

the most aggressive marine invaders (Norton 1976).

Common in shallow subtidal habitats, S. muticum is a

brown seaweed native to southeast Asia, having

become widely distributed as an invasive species

around Europe since the 1970s. Shipments of Japanese

oysters for aquaculture, together with accidental

transportation of fertile fronds by currents and boats,

are likely to have been the sources for introduction and

dispersal of this species (Critchley et al. 1986).

Outside its native range, S. muticum is invasive and

is able to tolerate a wide range of abiotic conditions

such as salinity or temperature (Norton 1976), making

it a strong competitor that can limit the distribution of

native species (Stæhr et al. 2000). Additionally, it has

rapid growth, high fecundity and is able to produce

fertile drifting fragments which constitute an efficient

mode of dispersal (Norton 1977; Pedersen et al. 2005).

These mechanisms enable S. muticum to easily

colonise and occupy new habitats, forming persistent

local populations (Arenas et al. 2002). Once estab-

lished, S. muticum can cause dramatic changes in

the sublittoral area it colonises, being commonly

2564 T. Salvaterra et al.

123

associated with the replacement of native species, an

increase of filamentous epiphytic algae and changes in

composition of flora and fauna (Critchley et al. 1986;

Sanchez and Fernandez 2005). Dense stands may

compete for space, increase sedimentation and reduce

nutrients, light or oxygen available to native algae,

which provide habitat and food to a wide variety of

marine fauna (Cacabelos et al. 2010). Literature

suggests that the impact of S. muticum on native

communities is highly variable, depending on the

habitat or species assemblages where it becomes

established (Viejo 1997; Stæhr et al. 2000; Britton-

Simmons 2004). Furthermore, Isbell et al. (2011)

suggest that many species are needed for ecosystem

functioning to be maintained at large spatiotemporal

scales, and that changes in composition of these

species are expected to decrease ecosystem process

rates. This highlights the need to examine its general

impacts on a wide range of ecosystem properties.

Here, a field-based experiment was established to

investigate the effects of S. muticum on the structure

and functioning of a benthic ecosystem, based on a

range of commonly occurring native algal assem-

blages. The following hypotheses were tested: (1) the

presence of S. muticum will impact negatively on

ecosystem functioning; (2) the presence of S. muticum

will reduce native faunal species diversity; (3) the

presence of S. muticum will alter the structure of the

food web; (4) the impact of S. muticum will differ

depending on whether or not its arrival is associated

with reductions in the density of native algal species;

(5) the impact of S. muticum will vary depending on

the native algal assemblage.

Materials and methods

Study site and algal survey

The experiment was conducted at Lough Hyne, a

marine reserve located in County Cork, southwest

Ireland (51�2905200N, 9�1704600W), where S. muticum

became established as an invasive species in 2003

(Simkanin 2004). This study site is a fully marine (at

34 %), yet sheltered sea lough (Kitching 1991), with

biological diversity and water temperature represen-

tative of the surrounding coastline (Rawlinson et al.

2004), making it an ideal location for field experi-

mentation (O’Gorman and Emmerson 2009). Before

setting up the experiment, the subtidal study site was

surveyed at a depth of 1–2 m at low tide by

snorkelling, to quantify the macroalgal community.

While S. muticum is prevalent in many locations at

Lough Hyne, we chose to survey the native algal

community in an area that was not impacted by the

invader. The survey was done using eight randomly

placed quadrats (0.25 m2), divided into 25 smaller

squares. A value from 0 to 100 % was attributed by

observation to the algae in each square. Mean

percentage cover for the entire quadrat was then

determined. The three most prevalent species were:

Cladostephus spongiosus (53 %), Fucus vesiculosus

(19 %) and Ceramium virgatum (3 % cover; see Table

S1). Fronds of these three macroalgae were then

collected from the same location of the survey.

Randomly selected fronds were weighed in order to

establish an average frond weight for each species

(Strong and Dring 2011).

Experimental design

The experiment was designed to manipulate two

factors: the invasive species and the native algal

assemblage to which it was added (see Fig. 1). Three

levels of invasive species were employed: (1) absence

of S. muticum; (2) substitution of half of the native

algal density with S. muticum; and (3) addition of S.

muticum to the native algal assemblage, i.e. we used a

combined additive and substitutive design, as recom-

mended by Griffen (2006), for detecting emergent

effects of multiple species in experimental ecology.

Three assemblages of native algae were established

according to the prevalence of each algal species at

Lough Hyne: (1) C. spongiosus only; (2) C. spongio-

sus and F. vesiculosus; and (3) C. spongiosus, F.

vesiculosus and C. virgatum. This methodology was

used in order to establish native algal communities

representative of the natural shoreline (similar to

Carey and Wahl 2010) and to simulate the progressive

loss of less dominant species from those communities.

The macroalgal assemblages were created by attach-

ing algal fronds to aluminium mesh squares of

17 9 14 cm. Garden wire was used to attach the algae

to the mesh. Each frond was selected to correspond to

the following weights: C. spongiosus (5.4 ± 0.1 g); F.

vesiculosus (13.2 ± 0.2 g); C. virgatum (9.4 ± 0.2 g);

and S. muticum (11.8 ± 0.2 g). These represented

average weights of fronds found naturally during the

Impacts of the invasive alga 2565

123

survey. The algae were distributed evenly on the

mesh in 3 rows of 4 fronds making a total of 12

fronds per mesh in the absence and substitution of

S. muticum treatments. In the additive design, 6

additional fronds of S. muticum (corresponding to

half of the total native density) were added to each

mesh, interspersed among the 3 rows of native

algae (see Table S2).

Each treatment was set up inside 42 9 41 9 10 cm

plastic mesh containers, with a 7 mm mesh size. Cable

ties were used to attach the mesh squares containing

the algal assemblages to the insides of the containers.

This kind of experimental procedure, using reattached

algae, has previously been employed successfully by

Arenas et al. (2006). In addition, sampling substrates

were used to assess the invertebrate community that

developed in the containers. Settlement panels

(10 9 10 cm PVC squares) were used to quantify

sessile species and nylon pot scourers (approximate

radius, 4 cm; approximate height, 2 cm) were used to

quantify mobile species within the mesocosms, with

one of each sampling substrate per cage. Both

settlement panels and pot scourers have previously

been used to measure the density of benthic inverte-

brates (O’Gorman et al. 2008, 2010).

Once the treatments were set up, the containers

were placed in the shallow subtidal (1–2 m depth at

low tide) and secured to the rocky substratum with

clean gravel, spread evenly across the bottom of each

container. The weight of the gravel was sufficient to

keep the containers in place during the experiment,

due to the highly sheltered nature of Lough Hyne.

Benthic species were free to recruit naturally. There

were four replicates of each treatment in the experi-

ment, yielding a total of 36 experimental units. The

units were arranged in a randomised block design,

with four rows of nine containers, each separated by at

least 1 m. The experiment ran for 6 weeks, from 25th

February to 8th April 2011.

Laboratory procedures

After 6 weeks, the containers were collected, except

for one lost replicate of the native algae only treatment

comprising all three species (see Fig. 1). The algae

attached to the mesh was removed and weighed. The

pot scourers were preserved in 70 % ethanol for later

sorting and identification of the fauna present. Settle-

ment panels were also collected, although no species

had settled during the experiment. A species of

Ectocarpus algae was found to be heavily fouling

the containers and the surrounding subtidal areas of

shoreline. This algae was also collected from each

container and weighed to determine whether it had an

effect on the experimental outcomes. Response vari-

ables for the experiment were: ecosystem functioning,

diversity and food web properties. These response

variables are described in detail below.

Fig. 1 Visualisation of the experimental design used in the

current study. Here, two factors were manipulated: the invasive

alga (S. muticum) and native algal assemblage. The invasive

alga was a fixed factor with three levels: (1) absent, (2) present in

an additive design (6 fronds of S. muticum added to the native

algae) and (3) present in a substitutive design (some of the native

algae replaced by S. muticum to maintain the same overall

number of fronds as used in the native algae only treatment).

Native algal assemblage was a random factor with three levels:

(1) C. spongiosus; (2) C. spongiosus and F. vesiculosus; and (3)

C. spongiosus, F. vesiculosus and C. virgatum. Note that this

factor is considered random because the assemblages used

represent a subset of all possible native algal assemblages. The

selected species identities represent dominance patterns in the

shallow subtidal of the study site (see Table S1). For a detailed

description of the number of algal fronds used in each treatment,

see Table S2. CS = Cladostephus spongiosus, FV = Fucus

vesiculosus, CV = Ceramium virgatum. *one replicate was not

recovered at the end of the experiment

2566 T. Salvaterra et al.

123

Ecosystem functioning

Ecosystem functioning was measured in terms of

primary production. This was quantified as the

percentage change in biomass of each of the three

native algal species manipulated in the experiment and

the biomass of the Ectocarpus sp. While biomass is not

directly equivalent to production, a large number of

studies use it as a surrogate measure (Hector et al.

1999; Scurlock et al. 2002), with surface chlorophyll

(i.e. biomass of phytoplankton) also found to be a

reliable estimate of primary production in marine

systems (Friedrichs et al. 2009).

Diversity

All the animals present in the samples were identified

to species level where possible. Taxon richness was

calculated as the number of species per container. The

Shannon-Wiener index was used as a measure of

species diversity and calculated as -Rpi*ln(pi), where

pi is the relative abundance of each species (i), divided

by the total abundance of each treatment. Total

abundance was calculated as the sum of all individuals

in each container.

Food web properties

Food webs are visualisations of trophic connections

between consumers and resources, and are commonly

used to illustrate the biological structure of an ecosys-

tem. Using a list of all the species identified at the end of

the experiment, it was possible to infer the food web

structure for each of the experimental containers. An

existing database of publications containing gut content

analyses of all the identified organisms was used to

assign a direct consumer-resource link between each

species (O’Gorman and Emmerson 2009; O’Gorman

et al. 2010). This database is summarised in Table S3,

which contains all the species identified in the exper-

iment, their feeding modes and reference to 152 studies

used to collate the feeding link information. Sub webs

for each mesocosm were drawn from this core list of

consumer-resource interactions, based on the species

present in each container. Food web properties such as

number of links, linkage density, connectance, mean

food chain length and the proportions of basal,

intermediate and top species could then be calculated

for each mesocosm.

Statistical analysis

Two-way ANOVAs were used to test the effects of the

invasive alga S. muticum (a fixed factor with three

levels: (1) additive, (2) substitutive and (3) absent) on

native algal assemblages (a random factor, orthogonal

to the first and with three levels: (1) C. spongiosus; (2)

C. spongiosus and F. vesiculosus; and (3) C. spong-

iosus, F. vesiculosus and C. virgatum) and their

associated faunal communities in terms of ecosystem

functioning (primary production), diversity (taxon

richness and Shannon-Weiner diversity) and food web

structure (number of links, linkage density, connec-

tance, food chain length and proportions of basal,

intermediate and top species). Analyses of changing

biomass of S muticum, F. vesiculosus and C. virgatum

as response variables had to be modified because these

species had been experimentally manipulated and so

did not arise in all treatments. Thus, in the analysis of

change in biomass of S. muticum as a response

variable, the invasive alga factor only had two levels

(additive and substitutive). For the analysis of change

in biomass of F. vesiculosus, the native algal assem-

blage factor only had two levels (C. spongiosus and F.

vesiculosus; C. spongiosus, F. vesiculosus and C.

virgatum) because F. vesiculosus was absent from the

assemblage containing only C. spongiosus. For the

analysis of change in biomass of C. virgatum, the

native algal assemblage factor would have had only

one level (C. spongiosus, F. vesiculosus and C.

virgatum) and so a one-way ANOVA was performed,

with only the invasive alga factor. Given this variation

in the numbers of levels and factors for some response

variables and the fact that one replicate was lost (see

Fig. 1), the degrees of freedom differed depending on

the response variable. Food web properties were

derived using specialized food web analysis packages

written for R. All statistical analyses were carried out

in R version 2.14.0.

Results

The biomass of S. muticum decreased in all containers

during the experiment, but this loss was consistent

across all treatments. Here, there was no significant

difference in the change in biomass of S. muticum with

any of the three native algal assemblages, irrespective

of whether it was added to or substituted for native

Impacts of the invasive alga 2567

123

biomass (Two-way ANOVA: p [ 0.39 for all terms;

Fig. 2).

Ecosystem functioning

There were significant effects of S. muticum on

primary productivity (biomass changes) in the native

algal community. There was an interaction between

the two experimental factors (invasive species and

native assemblage) for the change in biomass of C.

spongiosus (Two-way ANOVA: F4,26 = 4.68;

p = 0.006; Fig. 3a). Here, the addition of S. muticum

led to a significantly lower biomass of C. spongiosus in

native assemblages comprising C. spongiosus only or

C. spongiosus and F. vesiculosus. The opposite effect

was found in the native assemblage comprising all

three species, with the addition of S. muticum leading

to a significantly higher biomass of C. spongiosus than

in the native algae only treatment. For F. vesiculosus,

the greatest increase in biomass occurred in the native

algal assemblage (without S. muticum). There was a

significantly lower increase in the biomass of F.

vesiculosus when S. muticum was substituted for

native algal biomass and particularly after the addition

of S. muticum (Two-way ANOVA: F2,17 = 6.761;

p = 0.007; Fig. 3b). There was no significant effect of

S. muticum on the change in biomass of C. virgatum,

which declined in all treatments (One-way ANOVA:

p [ 0.70 for all terms; Fig. 3c). There was a signif-

icantly lower biomass of Ectocarpus sp. in native

algal assemblages with C. spongiosus only (Two-way

ANOVA: F2,26 = 8.520; p = 0.001) and in the native

versus invaded communities (Two-way ANOVA:

F2,26 = 4.392; p = 0.023). This appeared to be

largely driven by the reduced biomass of Ectocarpus

sp. in the native algae only treatment with just C.

spongiosus (see Fig. 3d).

Diversity

There were no significant effects of S. muticum on

faunal community diversity at the end of the exper-

iment. There was a clear tendency for reductions in

taxon richness (Fig. 4a) and Shannon-Wiener diver-

sity (Fig. 4b), but these effects were not significant

(Two-way ANOVA: p [ 0.26 for all terms). This is

most likely due to the absence of apparent differences

among invasion treatments for the native algal

assemblage comprising C. spongiosus and F. vesicu-

losus and the high degree of variation among plots.

Food web properties

Representative sub webs for the native algal assem-

blage containing all three species are shown in Fig. 5,

highlighting the complexity of these webs and the

tendency for some properties to vary as a result of

absence, substitution or addition of S. muticum.

Generalist top predators such as fish (Gobius paga-

nellus, Symphodus melops and Pomatoschistus pictus)

and prawns (Palaemon elegans and P. serratus)

dominated the S. muticum webs. Specialist predators

(Macropodia tenuirostris and Retusa truncatula) and

an increased number of primary consumers were more

prevalent when S. muticum was absent. Overall, the

number of links (Fig. 6a) and the linkage density

(Fig. 6b) of the food webs that developed by the end of

the experiment were unaffected by the addition or

substitution of S. muticum to the native algal commu-

nity (Two-way ANOVA: p [ 0.79 for all terms).

However, there was a significant increase in the

connectance of the food webs when S. muticum was

added to the native algal community. This effect was

consistent for all three native assemblages (Two-way

ANOVA: F2,26 = 4.891; p = 0.016; Fig. 6c). There

were no significant differences in the mean food chain

length of any of the treatments (Two-way ANOVA:

p [ 0.70 for all terms; Fig. 6d). There were no

significant main effects of native algal diversity or S.

Fig. 2 Change in biomass of S. muticum over the course of the

experiment. Light grey bars signify the substitution of native

algal biomass with S. muticum. Dark grey bars represent the

addition of S. muticum to the native algal biomass. CS = C.

spongiosus; FV = F. vesiculosus; CV = C. virgatum

2568 T. Salvaterra et al.

123

muticum on the proportion of basal, intermediate or

top species in the food webs that developed by the end

of the experiment (Two-way ANOVA: p [ 0.18 for

all terms). The substitution of S. muticum for the native

algal assemblage comprising all three species led to a

significant increase in the proportion of intermediate

(t4,26 = 1.997; p = 0.05; Fig. 7b) and decrease in

the proportion of top (t4,26 = -2.170; p = 0.039;

Fig. 3 Ecosystem functioning effects in the experiment,

represented as percentage change in biomass of a C. spongiosus,

b F. vesiculosus and c C. virgatum and d log10 (biomass of

Ectocarpus sp. ? 1). White bars indicate the native algae only

treatments (no S. muticum present). Light grey bars signify the

substitution of native algal density with S. muticum. Dark grey

bars represent the addition of S. muticum to the native algal

density. CS = C. spongiosus; FV = F. vesiculosus; CV = C.

virgatum. There are missing bars in panels (b) and (c) because

F. vesiculosus was not included in the native algal assemblage

with C. spongiosus only and C. virgatum was only included in

the three species assemblage (see experimental design in Fig. 1

and Table S2)

Fig. 4 Effect of the experimental treatments on the diversity of the benthic faunal community: a taxon richness and b Shannon

diversity. All other information as for Fig. 3

Impacts of the invasive alga 2569

123

Fig. 7c) species relative to the native algae only

treatment, however.

Discussion

The results of this study provide evidence for the

potential of S. muticum to induce important changes in

native communities where it becomes established.

Here, S. muticum had significant effects on the native

algal community and associated fauna from the

shallow subtidal of Lough Hyne marine reserve.

These effects were particularly evident for the two

most common algae on the south shoreline of the

Lough, C. spongiosus and F. vesiculosus, which

underwent contrasting changes in biomass over the

course of the experiment. In addition, there was an

alteration in the structure of the faunal benthic

community, with the arrival of several generalist

species in association with S. muticum and a redistri-

bution of species as top and intermediate predators.

This in turn led to a higher proportion of realised links

within these communities, highlighted by the

increased connectance of the webs. Such effects could

have important implications for the stability of the

system.

The presence of S. muticum had a negative impact

on ecosystem functioning, inducing significant reduc-

tions in primary productivity (see Fig. 3), measured as

the percentage change in biomass of each native algal

species (supporting hypothesis 1). The reduction in

biomass of F. vesiculosus, observed with both addition

and substitution of S. muticum, is consistent with the

results of Stæhr et al. (2000) in Denmark, and Viejo

(1997) in northern Spain. These authors reported that

the invasive S. muticum affected the local community

through competitive interactions with thick, slow

growing algae such as Fucus spp. (Stæhr et al.

2000), reducing their abundance. Similarly, Britton-

Simmons (2004) showed that competition with S.

muticum reduced the abundance of native canopy and

understory algae, suggesting these effects were caused

by competition for light. Indeed, other authors (San-

chez et al. 2005; Olabarria et al. 2009; Baer and

Stengel 2010) have demonstrated that S. muticum

reduces native algal abundance through shading,

particularly fucoids. It is also interesting to note that

S. muticum has previously been shown to have a

negative impact on the settlement of fucoids after it

has become established (J.N. Griffin unpublished).

This suggests that S. muticum not only has the ability

to reduce the biomass of established fucoid commu-

nities, but may also inhibit further recruitment to

restore their biomass.

Changes in the biomass of C. spongiosus were not

as consistent as those observed for F. vesiculosus (see

Fig. 3). S. muticum clearly had a negative impact on

the biomass of C. spongiosus in assemblages where it

was found on its own or in combination with just

F. vesiculosus. This result appears to be consistent

with the positive effect of removing S. muticum from

intertidal plots in a previous experiment (Sanchez

and Fernandez 2005). However, the reversal of this

Fig. 5 Representative food web for each of the treatments

containing C. spongiosus, F. vesiculosus and C. virgatum:

a native algae only treatment with no S. muticum added; b S.

muticum substituted for some native algae; c S. muticum added

to native algae. A selection of properties are also shown for each

food web: number of species (S), number of links (L),

connectance (C) and proportions of intermediate (%I) and top

(%T) species. A list of species making up these and all other

webs in the study can be found in Table S3, along with their

feeding modes and the source of literature used to compile the

trophic linkages between them

2570 T. Salvaterra et al.

123

negative effect in the native algal assemblage with all

three species appeared to be due to the poor perfor-

mance of C. spongiosus in combination with both F.

vesiculosus and C. virgatum. Perhaps C. spongiosus is

a poor competitor in diverse macroalgal assemblages.

Indeed, previous work has shown that C. spongiosus

may experience negative growth in high diversity

macroalgal assemblages, independent of grazing

activity (Engelen et al. 2011).

Although results revealed a decrease in the biomass

of S. muticum over the course of the experiment (see

Fig. 2), this effect was consistent across all treatments.

This indicates that the experimental factors were not

the source of the decline and, as such, it should not

alter the interpretation of the results. The poor

establishment of S. muticum may have been caused

by the experimental disturbance, i.e. harvesting of the

algal fronds from the subtidal and attaching them to

artificial mesh squares. Alternatively, the decrease in

S. muticum biomass might have been caused by

environmental or biological factors unfavourable to

this particular species, e.g. the emergence of Ectocar-

pus sp. (see Fig. 3d). Ectocarpus is a filamentous

macroalgae that presents rapid nutrient intake and fast

growth rate, associated with a low nutrient storage

capacity (Worm and Sommer 2000). The presence of

this alga can therefore be associated with increased

concentrations of nutrients. As nutrient supply is

spatially and temporally variable and occurs in

irregular pulses (Karez et al. 2004), a sudden rise in

the nutrient concentration at Lough Hyne might have

contributed to the increase in biomass of Ectocarpus

sp. This alga may then have out-competed S. muticum

for space and light (Jacobucci et al. 2008), acting as an

epiphyte and fouling the algal community at the

subtidal area under study. Support for this hypothesis

can be found in (Baer and Stengel 2010), where low

competitiveness of S. muticum due to colonization by

Plylaiella litorallis, also an epyphite from the order

Ectocarpales, was observed on the Irish west coast.

Contrary to what was expected (hypothesis 2), the

presence of S. muticum did not cause significant

changes in the diversity of the native faunal species

associated with the alga. The lack of an overall

significant effect on species richness and diversity

suggests that S. muticum only weakly impacts upon

native faunal biodiversity. Indeed, Viejo (1999)

showed that S. muticum had no effect on the compo-

sition of epifaunal communities associated with native

algae. Thomsen et al. (2006) also revealed a lack of

Fig. 6 Effect of the

experimental treatments on

the food web properties of

the mesocosm communities:

a number of links; b linkage

density; c connectance;

d mean food chain length.

All other information as for

Fig. 3

Impacts of the invasive alga 2571

123

correlation between grazer distribution and S. muti-

cum. Although other authors have found significant

differences in the constituent epifauna found on native

algae and S. muticum (Monteiro et al. 2009; Cacabelos

et al. 2010), Gestoso et al. (2010) observed that this is a

pattern with spatial and temporal variability in mac-

roalgal assemblages due to physical or biological

factors. This implies that, while species richness and

diversity of native assemblages may remain largely

unchanged, their composition may be different as a

result of the introduced species. If these changes in

composition result in functional feeding differences

(such as altered distribution of generalist versus

specialist consumers), this may lead to modified

consumer-resource interactions in the community.

Thus, it is important to consider the changes in food

web structure to more fully understand the impacts of

invasive species within a system.

There was, however, a clear tendency for the

greatest species richness and diversity of benthic fauna

to occur in the native algal communities, with a

decrease in both richness and diversity as a result of

the addition or substitution of S. muticum (see Fig. 4).

Previous studies have observed a similar trend

(Monteiro et al. 2009; Strong et al. 2009). One

possible explanation for this trend is that S. muticum

provides less canopy cover than the native algae,

leaving less habitat structure for invertebrate species

to occupy. A clear link between decreased habitat

complexity and low levels of richness or diversity has

previously been shown by Tilman (1999) and Duffy

(2003). Further work is needed to determine whether

this observed trend for declining diversity is in fact a

reliable pattern.

Increases in food web connectance and the propor-

tion of intermediate species and a decrease in the

proportion of top species were observed as a result of

the presence of S. muticum (supporting hypothesis 3).

Connectance is the fraction of all possible links that

are realised within a food web (Dunne et al. 2002) and

is considered to be a measure of food web complexity

(Lafferty et al. 2006). As such, greater connectance

means a greater proportion of realised links, which

implies that predators tend to be more generalist, i.e.

they consume a greater number of prey species. This is

underlined by the observed redistribution of species as

intermediate or top species (see Fig. 7). Here, the S.

muticum treatments tended to be dominated by large,

highly generalist predators such as fish (G. paganellus,

S. melops and P. pictus) and prawns (P. elegans and P.

serratus). These species may have been attracted to

the S. muticum treatments as a result of the lower

habitat complexity (caused by the reductions in

biomass of C. spongiosus and F. vesiculosus described

above and the structural simplicity, in terms of canopy

cover, of S. muticum), thus leading to increased

susceptibility of their prey. Due to their highly

generalist feeding, they displaced many of the smaller,

specialist species as top predators, reducing the

proportion of top species in these treatments and

increasing the proportion of intermediate species.

Their generalist diet also increased the overall

Fig. 7 Effect of the experimental treatments on the proportions

of a basal, b intermediate and c top species in the mesocosm

food webs. All other information as for Fig. 3

2572 T. Salvaterra et al.

123

connectance of the system (a similar finding to

O’Gorman et al. 2012).

High connectance has been associated with greater

robustness of a food web to secondary extinctions

(Dunne et al. 2002), i.e. consumers are less likely to go

extinct because they have a greater diversity of

resources available to them. This robustness in turn

facilitates a more stable web. However, the observed

trend towards decreasing species richness and diver-

sity may indicate a simplification of web structure,

despite its higher connectivity (as shown by O’Gor-

man et al. 2012). In contrast to the compositional

changes found here, however, Gestoso et al. (2010)

reported that no epifaunal species were exclusively

associated with S. muticum and native Bifurcaria

bifurcata algal habitats. Also, Viejo (1999) estimated

the impact of S. muticum invasion on mobile epifauna

by comparing patterns of abundance and species

distributions among S. muticum and indigenous

equivalents (Cystoseira nodicaulis and F. vesiculo-

sus), finding no major differences in mobile epibiota

associated with any of these algae. Lang and Busch-

baum (2010) also noted that there were no significant

effects of S. muticum on the algal or infaunal

community. As such, impacts of S. muticum may be

context dependent, making it difficult to formulate

generalisations to other systems or different species of

invasive algae.

Overall, the results indicate that the impact of S.

muticum depends on whether it replaces native

biomass or contributes to the total biomass of the

assemblage where it becomes established. Here, the

most significant alterations to the system were

detected when the invader was added to existing

native biomass (supporting hypothesis 4). Addition of

S. muticum led to an increase in the total density of

algae in the experimental containers. As algal fronds

were more densely packed, S. muticum may have had a

stronger direct influence on the native algae. For

example, influences of shading would have been

stronger and competition for nutrients within the

system would have been greater. Moreover, indirect

effects due to the associated grazers or benthic

invertebrates seeking shelter in the macroalgal canopy

may also have been enhanced due to the proximity of

neighbouring algal fronds.

There was no clear evidence in this experiment that

variation in the composition of the native algal

assemblages altered the impact of S. muticum, with

many of the described patterns occurring in all three

assemblages (in contravention of hypothesis 5).

Changes to the number of species, links and propor-

tions of intermediate and top species were clearest

when all three native algal species were present,

however, suggesting some beneficial effects of this

high diversity combination (see Figs. 4, 5, 6).

Limitations of the study and future research

There are a number of caveats that need to be

considered when interpreting the results of this

experiment. Practical limitations necessitated a short

term experiment, lasting just 6 weeks. Additionally,

the experiment was carried out in early spring when

the productivity of the system was particularly low

(Jassby et al. 2002). Future research should seek to

carry out long-term experiments with periodical

sampling, as time of year is likely to be an important

factor contributing to the diversity and composition of

epibiota inhabiting macroalgae. A longer term exper-

iment would also facilitate an analysis of results on the

function of algal life cycles and patterns of succession

(Sanchez and Fernandez 2005).

The food webs for each experimental mesocosm

and their associated properties are calculated from a

core web of consumer-resource interactions (see Table

S3). While many of the feeding links in this core web

are drawn from gut content analyses carried out at

Lough Hyne (O’Gorman et al. 2010), some are also

based on published studies spanning a large geograph-

ical range. Webs based on such literature may

overestimate the width of consumer diets (Hall and

Raffaelli 1997). However, any errors associated with

this overestimation should be consistent across all the

treatments in the experiment and are unlikely to

influence the overall conclusions of the study.

Conclusion

This study illustrates that the introduced alga S.

muticum can have important effects on native com-

munities, causing decreases in the biomass of two

prominent species of native algae and altering the

properties of associated food webs. These results,

combined with other studies on the impact of S.

muticum in native floral and faunal communities

(Viejo 1997; Stæhr et al. 2000; Britton-Simmons

Impacts of the invasive alga 2573

123

2004; Pedersen et al. 2005; Britton-Simmons et al.

2010), highlight the importance of research on the

impact of invasive algae. Due to their wide distribu-

tion and effects on native coastal communities,

introduced algae are potential agents of ecological

change. Moreover, this study reveals the importance

of a broader approach in the experimental methods

when it comes to invasive species. By only consider-

ing species composition or diversity, key changes to

underlying system properties may be missed. By

undertaking a food web approach, a redistribution of

species as top and intermediate predators was deter-

mined, with subsequent alterations to the connectivity

of the system. Thus, evaluating impacts on food web

properties is an important tool for estimating the

overall effects of invasive species. Considering the

biology of S. muticum, it is very likely that this alga

will continue to spread, expanding its range along the

coastlines of Europe. As such, future work should

build on the research undertaken here to fully under-

stand the potential system-level impacts of S. muticum

and to identify management strategies to control the

spread of this invasive alga and minimise its impact.

Acknowledgments We would like to thank Judith Kochmann,

Emmi Virkki and Jen Coughlan from the GROG group and

MarBEE lab at University College Dublin, for assistance with

field and laboratory work. We also thank Marina Cunha, Clara

Rodrigues and Guy Woodward for valuable comments on the

manuscript. TS was supported by an Erasmus training program

scholarship. DG was funded by the project SIMBIOSYS (2007-

B-CD-1-S1) as part of the Science, Technology, Research and

Innovation for the Environment (STRIVE) Programme,

financed by the Irish Government under the National

Development Plan 2007–2013, administered on behalf of the

Department of the Environment, Heritage and Local

Government by the Irish Environmental Protection Agency

(EPA). EOG is a Postdoctoral Research Fellow funded by

NERC (Grant NE/I009280/1) and was supported by the Irish

Research Council for Science Engineering and Technology’s

EMPOWER initiative during part of this study.

References

Arenas F, Sanchez I, Hawkins SJ, Jenkins SR (2006) The in-

vasibility of marine algal assemblages: role of functional

diversity and identity. Ecol 87:2851–2861

Arenas F, Viejo RM, Fernandez C (2002) Density-dependent

regulation in an invasive seaweed: responses at plant and

modular levels. J Ecol 90:820–829

Baer J, Stengel DB (2010) Variability in growth, development

and reproduction of the non-native seaweed Sargassum

muticum (Phaeophyceae) on the Irish west coast. Estuar

Coast Shelf Sci 90:185–194

Baiser B, Russell GJ, Lockwood JL (2010) Connectance

determines invasion success via trophic interactions in

model food webs. Oikos 119:1970–1976

Britton-Simmons KH (2004) Direct and indirect effects of the

introduced alga Sargassum muticum on benthic, subtidal

communities of Washington State, USA. Mar Ecol Prog

Ser 277:61–78

Britton-Simmons KH, Pister B, Sanchez I, Okamoto D (2010)

Response of a native, herbivorous snail to the introduced

seaweed Sargassum muticum. Hydrobiologia 14:324–334

Cacabelos E, Olabarria C, Troncoso JS (2010) Do grazers prefer

invasive seaweeds? J Exp Mar Biol Ecol 303:182–187

Carey MP, Wahl DH (2010) Native fish diversity alters the

effects of an invasive species on food webs. Ecology

91:2965–2974

Colautti RI, MacIsaac HJ (2004) A neutral terminology to define

‘invasive’ species. Divers Distrib 10:135–141

Critchley AT, Farnham WF, Morrell SL (1986) An account of

the attempted control of an introduced marine alga, Sar-

gassum muticum, in Southern England. Biol Conserv 35:

313–332

Crooks JA (2002) Characterizing ecosystem-level conse-

quences of biological invasions: the role of ecosystem

engineers. Oikos 97:153–166

de Wit MP, Crookes DJ, van Wilgen BW (2001) Conflicts of

interest in environmental management: estimating the

costs and benefits of a tree invasion. Biol Invasions 3:

167–178

Duffy JE (2003) Biodiversity loss, trophic skew and ecosystem

functioning. Ecol Lett 6:680–687

Dunne JA, Williams RJ, Martinez ND (2002) Network structure

and biodiversity loss in food webs: robustness increases

with connectance. Ecol Lett 5:558–567

Engelen A, Henriques N, Monteiro C, Santos R (2011) Mes-

ograzers prefer mostly native seaweeds over the invasive

brown seaweed Sargassum muticum. Hydrobiologia 669:

157–165

Fridley JD, Rl Brown, Bruno JF (2004) Null models of exotic

invasion and scale-dependent patterns of native and exotic

species richness. Ecology 85:3215–3222

Fridley JD, Stachowicz JJ, Naeem S, Sax DF, Seabloom EW,

Smith MD, Stohlgren TJ, Tilman D, Von Holle B (2007)

The invasion paradox: reconciling pattern and process in

species invasions. Ecology 88:3–17

Friedrichs MAM, Carr ME, Barber RT, Scardi M, Antoine D,

Armstrong RA, Asanuma I, Behrenfeld MJ, Buitenhuis ET,

Chai F, Christian JR, Ciotti AM, Doney SC, Dowell M,

Dunne J, Gentili B, Gregg W, Hoepffner N, Ishizaka J,

Kameda T, Lima I, Marra J, Melin F, Moore JK, Morel A,

O’Malley RT, O’Reilly J, Saba VS, Schmeltz M, Smyth TJ,

Tjiputra J, Waters K, Westberry TK, Winguth A (2009)

Assessing the uncertainties of model estimates of primary

productivity in the tropical Pacific Ocean. J Mar Syst

76:113–133

Gestoso I, Olabarria C, Troncoso JS (2010) Variability of epi-

faunal assemblages associated with native and invasive

macroalgae. Mar Freshw Res 61:724–731

Gribben PE, Wright JT (2006) Invasive seaweed enhances

recruitment of a native bivalve: roles of refuge from

2574 T. Salvaterra et al.

123

predation and the habitat choice of recruits. Mar Ecol Prog

Ser 318:177–185

Griffen BD (2006) Detecting emergent effects of multiple

predator species. Oecologia 148:702–709

Grosholz E (2002) Ecological and evolutionary consequences of

coastal invasions. Trends Ecol Evol 17:22–27

Hall SJ, Raffaelli DG (1997) Food web patterns: what do we

really know? In: Gange AC et al (eds) Mutitrophic inter-

actions. Blackwells Scientific Publications, Oxford,

pp 395–417

Hector A, Schmid B, Beierkuhnlein C, Caldeira MC, Diemer M,

Dimitrakopoulos PG, Finn JA, Freitas H, Giller PS, Good J,

Harris R, Hogberg P, Huss-Danell K, Joshi J, Jumpponen A,

Korner C, Leadley PW, Loreau M, Minns A, Mulder CPH,

O’Donovan G, Otway SJ, Pereira JS, Prinz A, Read DJ,

Scherer-Lorenzen M, Schulze ED, Siamantziouras ASD,

Spehn EM, Terry AC, Troumbis AY, Woodward FI, Yachi S,

Lawton JH (1999) Plant diversity and productivity experi-

ments in European grasslands. Science 286:1123–1127

Irigoyen AJ, Trobbiani G, Sgarlatta MP, Raffo MP (2011)

Effects of the alien algae Undaria pinnatifida (Phaeophy-

ceae, Laminariales) on the diversity and abundance of

benthic macrofauna in Golfo Nuevo (Patagonia, Argen-

tina): potential implications for local food webs. Biol

Invasions 13:1521–1532

Isbell F, Calcagno V, Hector A, Connolly J, Harpole WS, Reich

PB, Scherer-Lorenzen M, Schmid B, Tilman D, van Ruij-

ven J, Weigelt A, Wilsey BJ, Zavaleta ES, Loreau M

(2011) High plant diversity is needed to maintain ecosys-

tem services. Nature 477:199

Jacobucci GB, Guth AZ, Leite FPP (2008) Experimental eval-

uation of amphipod grazing over biomass of Sargassum

filipendula (Phaeophyta) and its dominant epiphyte. Nau-

plius 16:65–71

Jassby AD, Cloern JW, Cole BE (2002) Annual primary pro-

duction: patterns and mechanisms of change in a nutrient-

rich tidal ecosystem. Limnol Oceanogr 47:698–712

Karez R, Engelbert S, Kraufvelin P, Pedersen MF, Sommera U

(2004) Biomass response and changes in composition of

ephemeral macroalgal assemblages along an experimental

gradient of nutrient enrichment. Aquat Bot 78:103–117

Kennedy TA, Naeem S, Howe KM, Knops JMH, Tilman D,

Reich P (2002) Biodiversity as a barrier to ecological

invasion. Nature 417:636–638

Kitching JA (1991) Introduction to the ecology of Lough Hyne.

In: Myers AA, Little C, Costello MJ, Partidge JC (eds) The

ecology of Lough Hyne. Royal Irish Academy, Dublin,

pp 13–16

Lafferty KD, Dobson AP, Kuris AM (2006) Parasites dominate

food web links. Proc Natl Acad Sci USA 103:11211–11216

Lang AC, Buschbaum C (2010) Facilitative effects of intro-

duced Pacific oysters on native macroalgae are limited by a

secondary invader, the seaweed Sargassum muticum. J Sea

Res 63:119–128

Levin PS, Coyer JA, Petrik R, Good TP (2002) Community-

wide effects of nonindigenous species on temperate rocky

reefs. Ecology 83:3182–3193

Monteiro CA, Engelen AH, Santos ROP (2009) Macro- and

mesoherbivores prefer native seaweeds over the invasive

brown seaweed Sargassum muticum: a potential regulating

role on invasions. Mar Biol 156:2505–2515

Norton TA (1976) Why is Sargassum muticum so invasive? Br

Phycol J 11:197–198

Norton TA (1977) The growth and development of Sargassum

muticum (Yendo) Fensholt. J Exp Mar Biol Ecol 26:41–53

O’Connor NE, Crowe TP (2005) Biodiversity loss and ecosys-

tem functioning: distinguishing between number and

identity of species. Ecology 86:1783–1796

O’Gorman EJ, Emmerson MC (2009) Perturbations to trophic

interactions and the stability of complex food webs. Proc

Natl Acad Sci USA 106:13393–13398

O’Gorman EJ, Enright RA, Emmerson MC (2008) Predator

diversity enhances secondary production and decreases the

likelihood of trophic cascades. Oecologia 158:557–567

O’Gorman EJ, Jacob U, Jonsson T, Emmerson MC (2010)

Interaction strength, food web topology and the relative

importance of species in food webs. J Anim Ecol 79:

682–692

O’Gorman EJ, Fitch JE, Crowe TP (2012) Multiple anthropo-

genic stressors and the structural properties of food webs.

Ecology 93:441–448

Olabarria C, Rodil IF, Incera M, Troncoso JS (2009) Limited

impact of Sargassum muticum on native algal assemblages

from rocky intertidal shores. Marine Environ Res 67:

153–158

Pedersen MF, Stæhr PA, Wernberg T, Thomsen MS (2005)

Biomass dynamics of exotic Sargassum muticum and

native Halidrys siliquosa in Limfjorden, Denmark—

implications of species replacements on turnover rates.

Aquat Bot 83:31–47

Rapoport EH (1994) Remarks on marine and continental bio-

geography: an areographical viewpoint. Philos Trans Biol

Sci 343:71–78

Rawlinson KA, Davenport J, Barnes DKA (2004) Vertical

migration strategies with respect to advection and stratifi-cation in a semi-enclosed lough: a comparison of mero- and

holozooplankton. Mar Biol 144:935–946

Sanchez I, Fernandez C (2005) Impact of the invasive seaweed

Sargassum muticum (phaeophyta) on an intertidal macro-

algal assemblage. J Phycol 41:923–930

Sanchez I, Fernandez C, Arrontes J (2005) Long-term changes

in the structure of intertidal assemblages after invasion by

Sargassum muticum (Phaeophyta). J Phycol 41:942–949

Schaffelke B, Hewitt CL (2007) Impacts of introduced sea-

weeds. Bot Mar 50:397–417

Scheibling RE, Gagnon P (2006) Competitive interactions

between the invasive green alga Codium fragile ssp. to-

mentosoides and native canopy-forming seaweeds in Nova

Scotia (Canada). Mar Ecol Prog Ser 325:1–14

Scurlock JMO, Johnson K, Olson RJ (2002) Estimating net

primary productivity from grassland biomass dynamics

measurements. Global Change Biol 8:736–753

Simkanin CM (2004) The invasive seaweed Sargassum muti-

cum (Yendo) Fensholt in Lough Hyne Marine Nature

Reserve, Co Cork. Irish Nat J 27:481–482

Stachowicz JJ, Fried H, Osman RW, Whitlatch RB (2002)

Biodiversity, invasion resistance, and marine ecosystem

function: reconciling pattern and process. Ecology 83:

2575–2590

Stæhr PA, Pedersen MF, Thomsen MS, Wernberg T, Krause-

Jensen D (2000) Invasion of Sargassum muticum in

Limfjorden (Denmark) and its possible impact on the

Impacts of the invasive alga 2575

123

indigenous macroalgal community. Mar Ecol Prog Ser

207:79–88

Strong JA, Dring MJ (2011) Macroalgal competition and inva-

sive success: testing competition in mixed canopies of

Sargassum muticum and Saccharina latissima. Bot Mar

54:223–229

Strong JA, Maggs CA, Johnson MP (2009) The extent of grazing

release from epiphytism for Sargassum muticum (Phaeo-

phyceae) within the invaded range. J Mar Biol Assoc UK

82:303–314

Thomsen MS, McGlathery KJ, Tyler AC (2006) Macroalgal

distribution patterns in a shallow, soft-bottom Lagoon, with

emphasis on the nonnative Gracilaria vermiculophylla and

Codium fragile. Estuar Coasts 29:465–473

Thomsen MS, Wernberg T, Tuya F, Silliman BR (2009) Evi-

dence for impacts of nonindigenous macroalgae: a meta-

analysis of experimental field studies. J Phycol 45:812–819

Tilman D (1999) The ecological consequences of changes in

biodiversity: a search for general principles. Ecology

80:1455–1474

Valentine JF, Heck KL, Cinkovich AM (2002) Impacts of sea-

grass food webs on marine ecosystems:a need for a broader

perspective. Bull Mar Sci 71:1361–1368

Viejo RM (1997) The effects of colonization by Sargassum

muticum on tidepool macroalgal assemblages. J Mar Biol

Assoc UK 77:325–340

Viejo RM (1999) Mobile epifauna inhabiting the invasive Sar-

gassum muticum and two local seaweeds in northern Spain.

Aquat Bot 64:131–149

Worm B, Sommer U (2000) Rapid direct and indirect effects of a

single nutrient pulse in a seaweed-epiphyte-grazer system.

Mar Ecol Prog Ser 202:283–288

2576 T. Salvaterra et al.

123