ArticlePDF Available

Functional changes as indicators of trawling disturbance on a benthic community located in a fishing ground (NW Mediterranean Sea)

Authors:

Abstract and Figures

Trawling disturbs benthic communities, eliminating the most vulnerable organisms and modifying habitat structure. While the cumulative effects of disturbance resulting from commercial trawling activities are poorly understood, several studies suggest that chronically disturbed commu- nities are dominated by opportunistic organisms. This study focuses on changes in functional compo- nents of the benthic community occurring in muddy sediments in a NW Mediterranean trawling ground, including an area that has not been fished for 20 yr. In both disturbed and undisturbed areas, the overall benthic community from the fishing ground was dominated by burrowing epifaunal deposit feeders and predators, and deep burrowing infaunal deposit feeders. The fished area had a higher abundance of burrowing epifaunal scavengers and motile burrowing infauna, while the undisturbed area was characterised by higher abundance of surface infauna, epifaunal suspension feeders and predatory fish. This study clearly demonstrates that changes in the functional compo- nents of a benthic community can result from fishing in areas dominated by organisms not considered especially vulnerable to trawling activities. Thus, fisheries managers aiming to reduce ecosystem dis- turbance must consider the implications of trawling on the structure and functioning of all types of benthic communities.
Content may be subject to copyright.
MARINE ECOLOGY PROGRESS SERIES
Mar Ecol Prog Ser
Vol. 334: 117129, 2007
Published March 26
INTRODUCTION
Continental shelves have been intensively fished
worldwide over the last century, often resulting in the
overexploitation of commercial species and the impov-
erishment of benthic communities (de Groot 1984,
Dayton et al. 1995, Pauly et al. 1998). Bottom trawling
disturbs benthic communities. However, there is a poor
understanding of the degree to which frequency and
intensity of fishing disturbance to the seafloor can
chronically modify benthic communities (Thrush et al.
1995, Auster et al. 1996, Jennings & Kaiser 1998, Hall
1999, Norse & Watling 1999, Kaiser et al. 2000).
Fisheries management aiming to reduce anthro-
pogenic disturbance on ecosystems should consider
the implications of fishing activities for both the struc-
ture and functioning of marine communities. This
implies that there is a need to find appropriate indica-
tors of the community response to fishing disturbance
(Link et al. 2002, Bustos-Baez & Frid 2003, Frid et al.
2005, Piet & Jennings 2005). Changes in the functional
components of the communities represent the organ-
isms’ adaptations to the environment and their
response to stress; therefore, focusing on how commer-
cial trawling has modified the functional components
of the communities can clarify the consequences of this
type of fishing for benthic ecosystems. (Bremner et al.
2003, Bremner et al. 2005). Numerous trawling impact
studies demonstrate that the organisms’ responses to
disturbance depend on their biological traits. For
example, those organisms possessing traits considered
opportunistic and less affected by trawling activities
© Inter-Research 2007 · www.int-res.com*Email: sdejuan@icm.csic.es
Functional changes as indicators of trawling
disturbance on a benthic community located in a
fishing ground (NW Mediterranean Sea)
S. de Juan
1,
*
, S. F. Thrush
2
, M. Demestre
1
1
Institut de Ciències del Mar, Passeig Marítim de la Barceloneta 37-49, 08003 Barcelona, Spain
2
National Institute of Water and Atmospheric Research, PO Box 11-115, Hamilton, New Zealand
ABSTRACT: Trawling disturbs benthic communities, eliminating the most vulnerable organisms and
modifying habitat structure. While the cumulative effects of disturbance resulting from commercial
trawling activities are poorly understood, several studies suggest that chronically disturbed commu-
nities are dominated by opportunistic organisms. This study focuses on changes in functional compo-
nents of the benthic community occurring in muddy sediments in a NW Mediterranean trawling
ground, including an area that has not been fished for 20 yr. In both disturbed and undisturbed areas,
the overall benthic community from the fishing ground was dominated by burrowing epifaunal
deposit feeders and predators, and deep burrowing infaunal deposit feeders. The fished area had a
higher abundance of burrowing epifaunal scavengers and motile burrowing infauna, while the
undisturbed area was characterised by higher abundance of surface infauna, epifaunal suspension
feeders and predatory fish. This study clearly demonstrates that changes in the functional compo-
nents of a benthic community can result from fishing in areas dominated by organisms not considered
especially vulnerable to trawling activities. Thus, fisheries managers aiming to reduce ecosystem dis-
turbance must consider the implications of trawling on the structure and functioning of all types of
benthic communities.
KEY WORDS: Functional traits · Commercial fishing · Ecosystem-based management · Rarity
Resale or republication not permitted without written consent of the publisher
Mar Ecol Prog Ser 334: 117129, 2007
are likely to respond positively to disturbance events
(Kaiser & Spencer 1994, Dayton et al. 1995, Philippart
1998, Thrush et al. 1998, Auster & Langton 1999, Frid
2003, Blanchard et al. 2004).
In this study, several functional traits were selected
to represent different components of the organisms’
sensitivities to trawling disturbance. These traits were
considered important in determining both the
responses of benthos to fishing activities and the
potential for changes in ecosystem function. Firstly,
feeding type reflects the adaptation of the organisms to
the habitat, e.g. numerous studies have detected sig-
nificant increases of motile scavengers in recently
trawled areas (Kaiser & Spencer 1994, Collie et al.
1997, Ramsay et al. 1998, Demestre et al. 2000, Rumohr
& Kujawski 2000, Bozzano & Sarda 2002). Deposit
feeders can also be favoured as trawling increases the
availability of organic matter on the sediments (Frid et
al. 2000), whereas filter feeders are highly affected by
the increase of suspended sediment following trawling
(Caddy 1973). Secondly, the motilities of the organisms
and their positions in or on the sediments also influ-
ence their vulnerabilities to trawling. Sedentary sur-
face organisms are strongly affected by the trawling
gear (Bergman & Hup 1992, Thrush et al. 1995, Auster
et al. 1996, Wassenberg et al. 2002). Thirdly, external
protecting structures and the body designs of organ-
isms can also affect vulnerabilities. Species protected
with a hard shell, or vermiform organisms are consid-
ered less vulnerable to trawling impact (Bremner et al.
2003, Blanchard et al. 2004). Finally, other important
traits considered in this paper are the body sizes and
life spans of organisms. Previous impact studies have
observed a shift from large slow growing fauna to less
vulnerable smaller and faster growing organisms
(Philippart 1998, Roth & Wilson 1998, Kaiser et al.
2000, Rumohr & Kujawski 2000).
The combination of these functional traits allowed
estimation of both the vulnerabilties of organisms and
the resilience of populations impacted by fishing activ-
ities. For example, trawling generally favours less vul-
nerable motile scavenging fauna over highly vulnera-
ble surface sedentary taxa (McConnaughey et al. 2000,
Frid 2003). Population resilience depends on the vul-
nerabilities of the individuals along with life history
traits and reproductive strategies. Small and fast grow-
ing organisms are thus more resilient to disturbance.
Consequently, highly vulnerable organisms with low
population resilience are not expected to be abundant
in trawled areas.
The aim of this study was to investigate the func-
tional components of a benthic community subjected to
chronic fishing disturbance. The study area was
located in a northwestern Mediterranean commercial
fishing ground that is exploited by a large trawling
fleet operating with well-defined seasonal activity. We
compared chronically disturbed and undisturbed ben-
thic communities to obtain the response of functional
traits of the faunal assemblages to commercial trawling
activities. Our goal was to test predictions and estimate
how trawling alters the different functional compo-
nents of communities, thus improving our knowledge
of the ecological implications of fishing.
This study highlights an opportunity to investigate a
soft-bottom community from a commercial fishing
ground, considering both the large epifaunal and
smaller infaunal organisms. Moreover, we emphasise
the focus of this study on muddy soft bottoms, one of
the most characteristic habitats for species targeted by
trawling fisheries worldwide.
MATERIALS AND METHODS
Characteristics of the study area. The study was
conducted in the Catalan Sea located in the northwest-
118
Fig. 1. Location of the study area in the northwestern
Mediterranean Sea
de Juan et al.: Trawling disturbance in a benthic community
ern sector of the Mediterranean Sea (Fig. 1). The fish-
ing ground selected is on the continental shelf off the
Ebro Delta. It has a surface area of 400 km
2
and depths
between 30 and 80 m. The sediment consists of mud
(>95%) and has a median grain size of 2.7 to 4.6 µm.
This section of the continental shelf extends for 70 km
with a gentle slope, and has highly homogenous sedi-
ment and physical parameters (Díaz et al. 1990). The
benthic community is considered typical of continental
shelf muddy bottoms, with low diversity and high sec-
ondary production (Demestre 1986).
This study was focused on an otter-trawl fleet from
the port of Sant Carles de la Ràpita. With 59 vessels,
this is the largest trawling fleet in the region, operating
in several well defined fishing grounds (Demestre
2006). Previous studies in the area, as well as historical
catch records of the trawlers and daily records from the
local auction, permitted selection of a fishing ground
characterised by seasonal variation in the activity of
the fleet. Fishing fleet activity in this area varies from
high effort during autumn and winter (from September
to February: 7550 average fishing h mo
1
), to low effort
in spring and the beginning of summer (from March to
June: 5520 average fishing h mo
–1
), and a 2 mo fishing
closure in summer (July and August). The whole area
was surveyed using side scan sonar images that regis-
tered the trawl marks on the sea bed. These images
confirmed that the fishing ground was evenly trawled.
Furthermore, the selected fishing ground includes an
area of 2.7 km
2
that has remained undisturbed for 20 yr
due to the fragmented remains of an abandoned oil
platform. This portion of seafloor was used as a refer-
ence area (Demestre 2006).
Sampling cruises and processing of samples. Two
study sites were selected within the fishing ground: a
fished site and the undisturbed area as a reference. A
survey carried out in the fishing ground prior to this
study characterised the area and identified the benthic
community assemblages and environmental character-
istics at the 2 study sites. The sites had similar habitat
characteristics (Demestre 2006).
Samples of benthic fauna were collected from the
fished and reference sites on 7 experimental cruises
timed to encompass seasonal variability, as well as the
variations in fishing activity (27 to 30 June 2003, with
low fishing activity; 14 to 17 July 2003, 28 to 31 July
2003 and 19 to 22 August 2003, during fishing closure;
26 to 29 September 2003 and 14 to 17 November 2003,
with high fishing activity; and 18 to 21 June 2004, with
low fishing activity).
Epifaunal and infaunal organisms were the focus of
this study. The epifaunal fraction comprised the large
macrofauna living on the sediment surface or within
the first few cm of sediment depth. They were col-
lected with a surface dredge. Infaunal organisms com-
prised smaller macrofauna buried in the sediments.
They were collected with a grab. The epifauna was
sampled with a surface dredge, similar to a 2 m beam
trawl, composed of a 2 m × 40 cm iron-framed aperture,
a 40 mm mesh size and a 10 mm cod-end. The mini-
mum sample size to estimate the species richness was
determined (based on Sanchez et al. 1998). Accord-
ingly, we made 3 hauls of approximately 15 min dura-
tion at 3 knots h
–1
at each site. A total of 3 replicate epi-
faunal samples was collected randomly on each cruise
at both fished and reference sites. Epifaunal samples
were standardised to a surface area of 1000 m
2
, thus
taking into account variations in haul duration.
To survey the infaunal benthos, sediment samples
were obtained with a 0.1 m
2
Van Veen grab. Samples
were collected at 5 fixed Stns located in each of the 2
study sites. Five grab samples were collected randomly
at each station to determine minimum sample size for
species richness estimates (Demestre 2006). Sediment
samples were sieved over a 1 mm mesh. Epifaunal and
infaunal organisms were identified to the lowest taxo-
nomical level practical, and counted. The data
obtained with the epibenthic dredge and the infaunal
grab were analysed separately, as they sampled differ-
ent components of the benthic community.
Infaunal and epifaunal functional groups. The
analyses were based on groups of organisms with com-
mon functional attributes that were predicted to
respond in similar ways to fishing disturbance. The
data set was reduced to 13 epifaunal species and 25
infaunal taxa to focus the study on the organisms that
contributed most to the differences between the fished
and reference sites. These species were selected by
the SIMPER procedure (PRIMER, Clarke & Warwick
1994) as those that accounted for 90% of the dissimilar-
ity between fished and reference sites. These species
were also the most abundant in the community. The
epifaunal species selected contributed 90% to total
epifaunal abundance, and belonged to Bivalvia, Gas-
tropoda, Crustacea, Echinodermata and Osteichthya.
The selected infaunal taxa contributed 80% to overall
infaunal abundance, and comprised members of the
Polychaeta, Crustacea, Bivalvia, Gastropoda, Echino-
dermata and the Nemertini. Organisms contributing
<1% to overall abundance were aggregated in the
analysis into the rare species group.
A set of functional traits was assigned for every
selected species (Table 1). The species were classified
into different functional categories based on informa-
tion from a variety of literature sources and from spe-
cialist knowledge. Each functional category was pre-
dicted to show either a negative, neutral or positive
response to trawling activity. The predicted response
of each trait to trawling disturbance was formulated on
the basis of previous trawling impact studies con-
119
Mar Ecol Prog Ser 334: 117129, 2007
ducted elsewhere, and on the biology of the local
organisms.
The functional traits were split up in several cate-
gories. Feeding mode comprised 4 categories for epi-
fauna: filter feeders, deposit feeders, predators and
scavengers. For infauna, only 3 categories were used,
with predators and scavengers included in the same
group. Motility was described at the scale of the fish-
ing disturbance, and ranged from sedentary (unable
to avoid the disturbance) to highly motile organisms.
Categories for position on the sediment were allo-
cated differently for the infaunal and epifaunal organ-
isms. Epifauna were sampled with a surface dredge
that also collected organisms burrowing into the sedi-
ment. While grabs collected infaunal organisms
buried in the sediment, they also included organisms
living in the sediment-water interface. External pro-
tecting structures and the categories of body design
were considered important determinants of infaunal
vulnerability to fishing gear. The maximum sizes and
life spans were also included as important traits of
organisms. However life spans (and consequent
resilience against disturbance) were not included in
all the analyses, as this information was not available
for the entire species list; with few exceptions the
infaunal organisms are thought to have a life span of
12 yr. Two summary classifications, vulnerability and
resilience, were estimated after combining the differ-
ent functional traits.
Finally, a category ‘rarity’ was defined for those or-
ganisms that individually contributed <1% to overall
abundance. Rare species may provide valuable infor-
mation for identifying the functional traits highly af-
fected by fishing activities. However, variations in the
abundance of this group could not be tested by inferen-
tial statistical analysis, due to the high percentage of
zero abundance values. Therefore, conclusions drawn
from these data must be considered with caution.
Statistical analyses. The functional categories
(except resilience and life span) were transformed into
a list of codes summarising the biological characteris-
tics of each species (Table 1). The abundances of spe-
cies possessing the same code were summed (Table 2),
after which we tested for differences between commu-
nities from fished and reference sites (PRIMER statisti-
cal package, Clarke & Warwick 1994). Similarity
between each pair of samples was calculated using the
Bray-Curtis similarity index, after square root transfor-
mation of the data to reduce the influence of dominant
groups. A non-metric multidimensional scaling ordina-
tion (MDS) was developed based on the similarity
matrix. An ANOSIM test was used to test the similarity
between samples from fished and reference sites.
Finally, the data were analysed with the SIMPER pro-
cedure to determine which codes accounted for the
observed dissimilarities between samples.
Subsequently, epifaunal and infaunal species were
combined into groups with common functional attrib-
utes. In this way, the list of codes was reduced to 6 epi-
faunal and 9 infaunal functional groups (Table 3).
Abundances of the different functional groups were
analysed to compare fished and reference sites at dif-
ferent levels of fishing activity. The densities of the
functional groups were expected to either increase or
decrease with fishing activities, depending on the
functional traits.
A 2-way analysis of variance (ANOVA) was per-
formed to identify significant differences (p <0.05) in
the abundances of the functional groups between
fished and reference sites and among the different
cruises (each cruise represents a different level of fish-
ing activity). Site and cruise were analysed as fixed
factors. A Tukey post-hoc comparison test was applied
to detect significant pairwise differences among com-
binations of Site × Cruise. When the factor interaction
term was significant, a multiple comparison test based
on specific contrasts was used to determine signifi-
cance of the interaction between different levels of
each factor. The Kolmogorov-Smirnov test was used to
test the assumption of normality, and Levene’s test
checked the homogeneity of error variances. When the
normality assumptions were not met, a log
10
transfor-
120
Functional traits Codes Functional traits Codes
Motility Protection
Sedentary S Tube t
Low L None n
Medium M
Size
High H
<1 cm 1
Position 15 cm 4
Surface S 510 cm 6
Subsurface B >10 cm 9
Feeding type Life Span
Filter feeders F <1 yr
Deposit feeders D 13 yr
Predators P >3 yr
Scavengers S
Resilience
Vulnerability Low
Low L Medium –
Medium M High
High H
Body design
Vermiform V
Shell S
Scales A
Table 1. List of functional traits and their corresponding codes
used in the multivariate analysis. Codes for epifaunal data
include: motility, position, feeding, vulnerability and size.
Codes included for infaunal data: motility, position, feeding,
vulnerability, body design and protection. –: not coded as trait
not in multivariate analysis
de Juan et al.: Trawling disturbance in a benthic community
mation was applied. In 2 cases (Group
1e and Group 9i in Table 3) the data
were not normally distributed,
because the differences in abundance
between sites were very large and
bimodal; therefore, a Kruskal-Wallis
non-parametric test was used to detect
significant differences between sites.
Univariate analyses were conducted
with the statistical package S-plus
(Becker et al. 1988).
RESULTS
Organisms classified as having low
vulnerability were the most common
within the the epifauna, and were
slightly more abundant at the fished
site than at the reference site (Fig. 2a).
Moderately vulnerable species ac-
counted for most of remaining epi-
fauna at both sites. Highly vulnerable
organisms were extemely rare at the
fished site and present at low abun-
dances at the reference site (Fig. 2a).
Most infaunal species were charac-
terised as medium to low vulnerability
with no large differences between
sites (Fig. 2b). Highly vulnerable in-
fauna were uncommon at both sites
(Fig. 2b).
Multivariate pattern
Multivariate analysis showed that
epifaunal communities from reference
and fished sites were 37.5% dissimi-
lar. The MDS ordination of samples
(Fig. 3a) and ANOSIM test clearly dis-
tinguished fished and reference sites
with a stress level of 0.17 and an R
value of 0.91 (p = 0.001). The SIMPER
procedure highlighted the importance
of 3 functional codes in discriminating
the 2 sites (Table 2a). The fished site
was characterised by motile burrow-
ing predators and scavengers (codes
HBPM4 and MBSL9), whereas the reference site had
higher abundance of highly motile surface predators
and burrowing low motility filter feeders (codes
HSPM9 and LBFH6).
Multivariate analyses of the infaunal data showed
that reference and fished sites were 39.6% dissimilar.
The MDS ordination (Fig. 3b) and ANOSIM test
divided fished and reference samples into 2 well
defined groups with a 0.14 stress level and a 0.98 R
value (p = 0.001). The SIMPER procedure highlighted
5 functional codes as best discriminators of fished and
reference sites (Table 2b). The fished community was
121
(a) Av. N- Av. N-
Codes Epifaunal species fished reference Diss/SD
LBFH6 Acanthocardia echinata 0.1 2.98 *2.13*
HSPM9 Citharus linguatula 4.95 22.54 *1.60*
Lepidotrigla cavillone
Arnoglossus laterna
HBPM4 Lesuerogobius suerii 4.69 0.57 *1.52*
LBDL4 Nucula nucleus 7.46 9.81 1.32
MBSL9 Astropecten irregularis 73.98 57.09 1.31
SBDL4 Trachythyone tergestina 31.61 22.28 1.24
Alpheus glaber
MBPM4 Goneplax rhomboides 11.29 21.98 1.24
MSSL6 Bolinus brandaris 9.87 6.97 1.24
Pagurus excavatus
Liocarcinus depurator
(b) Av. N- Av. N-
Codes Infaunal species fished reference Diss/SD
LBPLVn Lumbrinereis sp. 153.18 41.17 *2.84*
Nemertini
LSPMVt Marphysa bellii 0.76 41.66 *2.18*
SBDLVn Aricidea sp. 147.41 341.03 1.7*
Levinsenia gracillis
Labidoplax digitata
LBDMVn Sternaspis scutata 98.47 31.94 1.7*
SSDMVt Ampharetidae 79.94 142.8 *1.44*
Prionospio sp.
Scolelepis cantabra
SSFHSn Thyasira flexuosa 22.29 18.4 1.36
SBDLVt Capitellidae 180.82 190.17 1.33
Cirratulidae
LSPLVn Glycera sp. 18.82 17.37 1.31
SSFHAt Ampelisca tenuicornis 22.92 58.80 1.29
Corophium rotundirostre
SBDMAn Apseudes spinosus 70.11 34.81 1.24
Apseudes latreillei
SSDMVn Flabelligeridae 36 37.26 1.16
Paralacydonia paradoxa
SBDMSn Mysella bidentata 19.24 14.8 1.12
Nucula nucleus
LSPMSn Hyala vitrea 19.47 16.34 0.93
MSFMAn Leucon mediterraneus 29.6 17.06 0.83
LSDLAn Amphiura chiajei 0.94 3.14 0.77
Table 2. Results of SIMPER analysis for (a) epifaunal and (b) infaunal data.
Codes and corresponding species; average abundance at fished (Av. N-
fished) and reference (Av. N-reference) sites; index of contribution to dissimi-
larity between sites (Diss/SD); *indicates indices >1.40
Mar Ecol Prog Ser 334: 117129, 2007
characterised by vermiform motile burrowing preda-
tors and deposit feeders, which have low or moderately
vulnerabilities to trawling (codes LBPLVn and LBD-
MVn). The infauna from the reference site was charac-
terised by sedentary burrowing deposit feeders and
sedentary or low motility surface organisms living in
tubes (codes SBDLVn, SSDMVt, and LSPMVt).
Responses of epifaunal functional groups to
fishing activity
The abundances of epifaunal functional groups
(Table 3a) were analysed to test for significant differ-
ences between fished and reference sites and among
the different cruises (Fig. 4). The predicted and ob-
served response of these groups to trawling is included
in Table 4a.
The low motility burrowing filter feeders (Group 1e)
were characterised by long-life span and low popula-
tion resilience. As predicted, these organisms were
mostly found in low abundance at the reference site
122
(a) Epifaunal taxon Motil. Posit. Feed. Size Age Resil. Vuln. E O
Group
1e A. echinata LSSFF510>3 LH↓↓ ↓↓
2e N. nucleus, A. glaber, T. tergestina S/L SS DF <5 nd nd L–M ↑↑
3e C. linguatula, L. cavillone, A. laterna HSP>10>3MM↓↓
4e G. rhomboides, L. suerii M/H SS P <5 >2 nd M
5e B. brandaris, L. depurator, P. excavatus MS S>10>2HL↑↑
6e A. irregularis MSS S>10>3 H L ↑↑ ↑↑
(b) Infaunal taxon Motil. Posit. Feed. Shape Age Vuln. E O
Group
1i T. flexuosa, A. tenuicornis, C. rotundirostre S S FF Sh/Sc nd H ↓↓
2i L. mediterraneus M S FF Sc nd M
3i Ampharetidae, S. cantabra, Prionospio sp., S S DF Verm nd M
P. paradoxa, Flabillegeridae
4i A. chiajei M S DF Sc >1yr L
5i Capitellidae, L. gracilis, Aricidea sp., S SS DF Verm nd L
Cirratulidae, L. digitata
6i S. scutata M SS DF Verm >1yr M ↑↑
7i A. spinosus, A. latreillei, M. bidentata, S SS DF Sh/Sce >1yr M ↓↑
Nucula sp.
8i Glycera sp., M. bellii M S P Verm nd L–M ↓↓
9i H. vitrea, Nemertini, Lumbrinereis sp. M SS P Verm/Sh nd L–M ↑↑ ↑↑
Table 3. Epifaunal (a) and infaunal (b) functional groups. Functional group and species included in each group are indicated
in the first 2 columns (left side); the subsequent columns summarise the biological traits. Motility (Motil.): S, sedentary; L, low; M,
medium; H, high. Position on the substratum (Posit.): S, surface and SS, subsurface. Feeding mode (Feed.): FF, filter feeders; DF,
deposit feeders; P, predators; S, scavengers. Maximum size (Size) for epifauna (a) in cm. Body design (Shape) for infauna (b): Sh,
shell; Sc, scales; Verm, vermiform. Maximum age (Age) in yr. Resilience (Resil.) for the epifauna (a), and vulnerability (Vuln.) for
epifauna and infauna (a,b): L, low; M, medium; H, high. The last 2 columns (right side) indicate the expected (E) and observed
(O) response to trawling disturbance: , increase in abundance; , decrease in abundance; –, no predicted/observed response.
d: no data available
Fig. 2. Vulnerabilities of epifauna (a) and infauna (b) at fished
and reference sites
de Juan et al.: Trawling disturbance in a benthic community
(i.e. 25 ind. 1000 m
–2
; p < 0.001).
Small burrowing deposit feeders with
low motility (Group 2e) comprised the
second most dominant group in abun-
dance, with no important differences
detected between fished and refer-
ence sites (i.e. 1560 ind. 1000 m
–2
on
average at fished and reference sites).
The abundance decreased at both
sites in September and November,
during high fishing activity (p <
0.001). Most large surface organisms
with long life spans (Group 3e) were
highly motile predatory fish. As pre-
dicted, these organisms were signifi-
cantly more abundant at the reference
site (i.e. 210 and 1540 ind. 1000 m
–2
at fished site and reference sites, re-
spectively; p < 0.001). Fished site
abundance remained low throughout
the study, except for a transient in-
crease at the beginning of the fishing
closure (June to July 2003; p < 0.001).
The predatory organisms (Group 4e)
were characterised by small size,
motility and burrowing behaviour.
However, these organisms live >2 yr
and were classified as moderately vul-
nerable. The number of individuals
from this group was similar at refer-
ence and fished sites (i.e. average
545 ind. 1000 m
–2
). Abundance de-
creased in September and November,
associated with high fishing activity (p
< 0.001). Motile scavengers (Group
5e), characterised by surface position,
medium size and life spans >2 yr,
were highly resilient. These organ-
isms decreased at the fished site in
September and November cruises, co-
inciding with high fishing effort (i.e.
123
a
Stress: 0.17
b
Stress: 0.14
Fig. 3. Multidimensional scaling ordination plot of epifaunal (a) and infaunal (b) functional code abundances. M: correspond to
samples from fished site and
ds: from reference site
-2
0
2
4
6
8
Abundance (N 1000 m
–2
)
Fished site
Reference site
Group 1 - Epifauna
0
20
40
60
80
Group 2 - Epifauna
0
20
40
60
80
Group 3 - Epifauna
0
10
20
30
40
50
60
Group 4 - Epifauna
June '03
July '03
July '03
Aug. '03
Sept. '03
Nov. '03
June '04
-4
0
4
8
12
16
Group 5 - Epifauna
June '03
July '03
July '03
Aug. '03
S
ept. '03
Nov. '03
June '04
0
40
80
120
160
Group 6 - Epifauna
Fig. 4. Mean abundances (±SD) of epifaunal functional groups at fished and
reference sites in survey cruises from June 2003 to June 2004. See Table 3 (a) for
details of the 6 groups
Mar Ecol Prog Ser 334: 117129, 2007
510 ind. 1000 m
–2
in JuneAugust and 1 ind. 1000 m
–2
in September– November at the fished site; p < 0.001).
Reference site abundance decreased in November (p <
0.001). The low vulnerability burrowing scavengers
with large size and long life span (Group 6e) had high
population resilience. This was the most abundant
group in the community (i.e. 40100 ind. 1000 m
–2
aver-
age at fished and reference sites), with significantly
higher number of organisms at the fished site (p =
0.022).
Rare epifaunal organisms (1% of the overall abun-
dance) included sedentary or sessile surface filter feed-
ers (the Cnidarians Alcyonium palmatum and Veritil-
lum cynomorium), and fragile large deposit feeders,
represented by the sea urchins Schizaster canalifera
and Brissopsis atlantica.
Responses of infaunal functional groups to fishing
activity
Abundances of the infaunal functional groups
(Table 3b) were analysed to test for significant differ-
ences between fished and reference sites and between
cruises (Fig. 5). Table 4b includes the predicted and
observed response of infaunal functional groups to
trawling disturbance.
The highly vulnerable organisms, represented by
sedentary surface filter feeders with shells or scales
(Group 1i), were generally less abundant at the fished
site (i.e. 1580 and 50120 ind. m
–2
at the fished and
reference sites, respectively; p < 0.001). The abun-
dance followed a seasonal pattern at the reference
site, decreasing at the end of summer and autumn (p
< 0.001). Fished site abundance increased during the
fishing closure and during the high fishing activity
cruise in September (p < 0.001). The low abundance
and high variability between cruises hindered detec-
tion of significant patterns by analysis of variance of
the high motility taxa (Group 2i). Vermiform surface
deposit feeders of moderate vulnerability (Group 3i)
were significantly more abundant at the reference site
in the June and July cruises (p < 0.001). The abun-
dance decreased in August and increased again in
November (i.e. 100250 ind. m
–2
at reference site, and
60180 ind. m
–2
at fished site; p < 0.001). The density
of low vulnerability motile surface deposit feeders
(Group 4i) was low and patchy, which again hindered
the detection of significant patterns by the analysis of
variance. Sedentary vermiform subsurface deposit
feeders (Group 5i) were the most abundant organisms
in the community (i.e. 250750 ind. m
–2
average at
fished and reference sites) and exhibited significantly
higher densities at the reference site in the June and
July cruises (p < 0.001); densities decreased at the end
124
(a) Epifauna df F value Pr(F)
Group 2e
Site 1 3.99 0.056
Cruise 6 8.65 <0.001*
Site × Cruise 6 2.32 0.061
Group 3e
Site 1 147.56 <0.001*
Cruise 6 5.96 <0.001*
Site × Cruise 6 2.94 0.023*
Group 4e
Site 1 3.78 0.062
Cruise 6 3.36 <0.001*
Site × Cruise 6 2.15 0.078
Group 5e
Site 1 2.91 0.099
Cruise 6 3.75 0.007*
Site × Cruise 6 6.99 <0.001*
Group 6e
Site 1 5.86 0.022*
Cruise 6 5.21 0.001*
Site × Cruise 6 1.27 0.301
(b) Infauna df F value Pr(F)
Group 1i
Site 1 70.05 <0.001*
Cruise 6 18.78 <0.001*
Site × Cruise 6 12.25 <0.001*
Group 3i
Site 1 32.48 <0.001*
Cruise 6 11.10 <0.001*
Site × Cruise 6 2.99 0.013*
Group 5i
Site 1 43.81 <0.001*
Cruise 6 3.49 0.005*
Site × Cruise 6 3.33 0.007*
Group 6i
Site 1 147.02 <0.001*
Cruise 6 1.86 0.104
Site × Cruise 6 1.76 0.125
Group 7i
Site 1 53.42 <0.001*
Cruise 6 2.39 0.040*
Site × Cruise 6 4.59 <0.001*
Group 8i
Site 1 197.43 <0.001*
Cruise 6 5.02 <0.001*
Site × Cruise 6 1.47 0.207
(c) Kruskal-Wallis df χ
2
p
Group 1e
Site 1 325.92 <0.001*
Group 9i
Site 1 44.48 <0.001*
Table 4. ANOVA test summary. Epifaunal functional groups
analysis (a); infaunal functional groups analysis (b); Kruskal-
Wallis non-parametric test (c). Groups 2i and 4i were ex-
cluded from ANOVA due to low abundance and high vari-
ability recorded; see ‘Results’ for further details. *indicates
significant p-value (p < 0.05)
de Juan et al.: Trawling disturbance in a benthic community
of the fishing closure (p = 0.005). At the fished site this
variability was not detected. Motile subsurface
deposit feeders, with long life span and moderate vul-
nerability to trawling (Group 6i), were significantly
more abundant at the fished site (p < 0.001), and the
number of individuals was constant during the study
period (i.e. 65120 ind. m
–2
at fished site and 2040
ind. m
–2
at the reference site). Sedentary subsurface
deposit feeders, with shell or scales body design and
life span longer than one year (Group 7i), although
predicted to be vulnerable to trawling, were more
abundant at the fished site (i.e. 50120 ind. m
–2
at
fished site and 3070 ind. m
–2
at reference site; p <
0.001). Motile surface carnivores (Group 8i) were sig-
nificantly more abundant at the reference site (i.e.
5070 ind. m
–2
at reference, and 1025 ind. m
–2
at
fished site; p < 0.001), decreasing at both sites in
November and June 2004 cruises (p < 0.001). As pre-
dicted, burrowing carnivores (Group 9i) were signifi-
cantly more abundant at the fished site (140200 ind.
m
–2
at fished and 4090 ind. m
–2
at reference site; p <
0.001).
Rare infauna included the tube dwelling polychaetes
(Pectinaridae, Terebellidae and Serpulidae). These
organisms were found mostly at the reference site.
DISCUSSION
Responses of functional groups to fishing activity
Trawling disturbance modified the functional com-
ponents of the benthic community from the Sant Carles
de la Ràpita fishing ground. Functional group analyses
125
0
40
80
120
160
Fished site
Reference site
Group 1 - Infauna
0
40
80
120
160
Group 2 - Infauna
0
100
200
300
Group 3 - Infauna
-4
0
4
8
12
16
Group 4 - Infauna
0
200
400
600
800
1000
Group 5 - Infauna
0
40
80
120
160
200
Group 6 - Infauna
June '03
July '03
July '03
Aug. '03
Sept. '03
Nov. '03
June '04
0
40
80
120
160
200
Group 7 - Infauna
June '03
July '03
July '03
Aug. '03
Sept. '03
Nov. '03
June '04
0
20
40
60
80
100
120
Group 8 - Infauna
June '03
July '03
July '03
Aug. '03
Sept. '03
Nov. '03
June '04
0
50
100
150
200
250
Group 9 - Infauna
Abundance (N 1 m
–2
)
Fig. 5. Mean abundances (±SD) of infaunal functional groups at fished and reference sites in survey cruises from June 2003 to
June 2004. See Table 3 (b) for details of the 9 groups
Mar Ecol Prog Ser 334: 117129, 2007
differentiated infaunal and epibenthic communities in
the fished and reference sites, confirming the sensitiv-
ity of the functional traits to trawling disturbance and
the potential of benthic communities for functional
changes, even when species richness is low. Patterns
were apparent in spite of the dominance of organisms
with low or medium vulnerability to trawling. The most
abundant functional groups occurred at equal densi-
ties in the fished and reference sites. The epifaunal
community in the fishing ground had a high abun-
dance of small deposit feeders and predators, which
avoid direct contact with the fishing gear by burrowing
into the sediments. Sedentary deep burrowing deposit
feeders dominated the infaunal community in the fish-
ing ground. However it is difficult to ascertain if this
assemblage configuration results solely from the trawl-
ing disturbance history in the fishing ground, or if
it also represents a characteristic community from
muddy bottoms.
Muddy bottom communities from continental
shelves have been characterised previously by a pre-
dominance of subsurface deposit feeders (Roth & Wil-
son 1998, Pearson 2001, Pranovi et al. 2005). However,
the long history of exploitation of Mediterranean ben-
thic communities might have selected those less vul-
nerable organisms adapted to frequent anthropogenic
disturbance (Dayton et al. 1995, Tuck et al. 1998,
Thrush & Dayton 2002). The lack of pristine areas with
similar characteristics to the fishing ground hampers
our ability to conclusively attribute differences in
assemblages to chronic trawling impact. Nevertheless,
the comparison of fished and reference areas high-
lighted other functional groups showing a positive
response to trawling disturbance. Motile burrowing
scavengers dominated the epifaunal community at
both sites, but they were significantly more abundant
at the fished site. These organisms have low vulnera-
bility to trawling and benefit from the carrion supply in
the trawled area (Kaiser & Spencer 1994, Ramsay et al.
1998, Demestre et al. 2000). Importantly, our study
demonstrates that the increase in scavenger abun-
dance can be more than a transient response. Among
the infaunal organisms, the motile burrowing deposit
feeders and carnivores were significantly more abun-
dant at the fished site. Motility and burrowing behav-
iour proved to be important traits in determining vul-
nerability, as organisms that burrow deeper than the
penetration depth of the trawl gear may avoid trawling
disturbance (Philippart 1998, Ramsay et al. 1998,
Brown et al. 2005, Frid et al. 2005). These motile
deposit feeders and carnivores can therefore benefit
from the increase of organic matter available in the
sediment column in trawled areas (Frid et al. 2000).
Motile predators and sedentary filter feeders distin-
guished the undisturbed epifaunal community, while
the infaunal community was differentiated by a
higher abundance of surface organisms. This undis-
turbed area has not been trawled for 20 yr, suggesting
that these organisms are sensitive to fishing activities.
The highly vulnerable filter feeding epifauna were
significantly more abundant in the undisturbed area,
with densities in the fished site of <2 ind. 1000 m
–2
.
This group is represented by large and slow growing
sedentary organisms predicted to have little resilience
to trawling activities (Hill et al. 1999, Jennings et al.
2001, Blanchard et al. 2004). Filter feeding sedentary
infauna living on surface sediments also proved vul-
nerable to fishing activities, as significantly higher
densities of these organisms were recorded in the
undisturbed area. These results agree with predic-
tions that filter feeders can be highly affected by
trawling, due to increased suspended sediment con-
centrations (Caddy 1973). The undisturbed infaunal
community also had significantly higher abundances
of motile carnivores and vermiform sedentary deposit
feeders living within surface sediments. When taking
into consideration their biological traits, these organ-
isms were expected to be less vulnerable to trawling,
however, their surface position increases vulnerability
to fishing disturbance (Bergman & Hup 1992). The
large motile predators were also significantly more
abundant in the undisturbed epifaunal community.
These results confirm the suggestion that some fish
species are driven out of heavily fished areas, which
become dominated by more opportunistic scavengers
(Collie et al. 1997, Bozzano & Sarda 2002, Blanchard
et al. 2004, Daan et al. 2005). However, it is important
to note that the fish driven out of heavily fished area
are the target species of the trawling fleet. Regardless
of the mechanism that reduces populations of large
motile fish predators in the fishing grounds, the
higher abundance of these organisms in the undis-
turbed area suggests closed areas are effective as
refuges for commercial species.
Seasonality vs. fishing intensity
The seasonality observed in both epifaunal and
infaunal communities matches the description by
Sardá et al. (1999) of Mediterranean community dy-
namics, with a decrease in faunal abundance at the
end of summer and during autumn. The most abun-
dant epifaunal organisms followed a seasonal pattern
marked by a decrease in the numbers of individuals in
autumn (September and November). The burrowing
scavengers (opportunistic and dominant in the com-
munity) did not follow this seasonal trend. Infaunal
organisms also showed a seasonal pattern, with higher
abundance in spring and early summer, followed by a
126
de Juan et al.: Trawling disturbance in a benthic community
decrease in late summer and autumn (August and Sep-
tember). Sometimes, a second peak of abundance
occurred in November. In the case of the infaunal com-
munity, this seasonal pattern was generally observed
only at the undisturbed site.
These results confirm the negative effects of trawl-
ing, as the lack of seasonality can be linked to the elim-
ination of natural variation by disturbance events (Hall
1999). The fishing ground surveyed was characterised
by variability in the trawling fleet activity regimen:
high fishing effort from September to February, low
effort from March to June, and a 2 mo fishing closure
in July and August. The decrease in the number of epi-
faunal organisms in September and November coin-
cided with the highest fishing effort in the area. Unfor-
tunately, as the number of individuals decreased
equally at both sites, no correlation with the fishing
activity can be established.
The reference site corresponds to a small portion of
the fishing ground that remains un-trawled, but it is
otherwise surrounded by intense trawling activity.
Therefore, it is difficult to eliminate indirect influences
of trawling activity, operating through effects on the
supply of colonists or increases in turbidity (Hill et al.
1999, Palanques et al. 2001). The close correlation
between the fishing regimen and biological seasonal-
ity imposes a further difficulty in comparing the fished
and reference areas. Low fishing activity and the fish-
ing closure occur during spring and summer, whereas
high fishing activity is undertaken in autumn months.
Mediterranean communities are characterised by an
increase of the number of individuals in spring and a
sharp decrease at the end of summer and autumn,
coinciding with the highest levels of fishing activity.
However, some organisms showed specific temporal
responses to trawling activities in the study area. The
motile surface epifaunal scavengers underwent a
sharp decrease of abundance during the period of high
fishing effort in the fished area. In contrast, the less
vulnerable motile burrowing scavengers were abun-
dant even during the high effort season. No infaunal
organisms showed a significant response to the tempo-
ral changes in fishing activity regimen. Intense fishing
in this portion of the continental shelf might have
resulted in adapted communities less sensitive to the
activity levels (Thrush et al. 1995, Sanchez et al. 1998,
Auster & Langton 1999, McConnaughey et al. 2000).
Functional traits shift: relevance to community
function
Burrowing scavengers dominated the epifaunal
community, and responsed positively to fishing activi-
ties. The infaunal community was dominated by deep
burrowing deposit feeders, and at the fished site,
which was characterised by a lack of seasonality, there
was a higher abundance of motile burrowing organ-
isms. The undisturbed site contained higher functional
diversity, with more epifaunal sedentary suspension
feeders and large predatory fish, and a higher number
of surface infauna living within tubes. The results sup-
port the predictions that opportunistic traits are benefi-
cial in disturbed areas (Dayton et al. 1995, Bremner et
al. 2005). In contrast, those traits predicted to be more
vulnerable to fishing activities were more abundant in
the undisturbed area, presumably because the organ-
isms with these functional characteristics were unable
to withstand frequent trawling disturbance. The
organisms that respond negatively to trawling distur-
bance can be considered better indicators than oppor-
tunistic species, which tend to adapt easily and
respond weakly to disturbance (Frid 2003). The fragile
sea urchins Schizaster canaliferus and Brissopsis
atlantica, several sessile surface filter feeders (e.g. the
Cnidarian Alcyonium palmatum), and surface sessile
polychaetes living in tubes were rare organisms in the
fishing ground. These organisms are considered
highly vulnerable to trawling, and considerably re-
duced in abundance by fishing activities (de Groot
1984, Kaiser & Spencer 1994, Hill et al. 1999, Kaiser et
al. 2000, Wassenberg et al. 2002). Focusing on these
organisms, we gain knowledge about the most sensi-
tive functional traits, which may have been eliminated
from heavily fished continental shelves. However,
despite their low abundance these organisms can play
an important role in ecosystem functioning, and
increase the functional diversity of the community,
which is important for ecosystem resilience.
The dominance of opportunistic traits in the benthic
community from the fishing ground might have led to
an alternative stable state, with organisms adapted to a
frequent disturbance regimen (Auster & Langton
1999). Nevertheless, it is difficult to know the larger
implications of the shift in functional traits for ecosys-
tem functioning (Thrush & Dayton 2002). Focusing on
the response of those traits that proved to be sensitive
to trawling activities might help in understanding
changes observed in chronically impacted communi-
ties. Functional traits reflect organismhabitat rela-
tionship and the ability of species to endure distur-
bance events or to move (Fauchald & Jumars 1979,
Roth & Wilson 1998, Pearson 2001, Frid et al. 2005). In
our study, scavengers appeared to replace the less
opportunistic predators. This could have important
implications for overall benthic community composi-
tion and alter predatorprey relationships (Bozzano &
Sarda 2002, Thrush & Dayton 2002, Piet & Jennings
2005). Predatory fish are a target for the trawling fleet,
and the higher abundance of these organisms in the
127
Mar Ecol Prog Ser 334: 117129, 2007
undisturbed area may indicate that this area better ful-
fils the habitat requirements of these organisms. Frag-
ile organisms like sea urchins, or large sedentary filter
feeders, although generally at low densities, also
appear to have been replaced by burrowing motile
epifauna. The elimination of such large organisms
leads to reduction of habitat structure, with implica-
tions for community diversity. Moreover, these organ-
isms can have important roles in bioturbation and
nutrient flux at the sedimentwater interface (Dayton
et al. 1995, Thrush et al. 1995, Auster et al. 1996, Wid-
dicombe et al. 2000, Wassenberg et al. 2002, Lohrer et
al. 2004). The dominance of relatively small organisms
is probably an adaptation to frequent trawling distur-
bance, which implies the elimination of large slow
growing organisms (Philippart 1998, Roth & Wilson
1998, Rumohr & Kujawski 2000). The infaunal commu-
nity from the fished area was characterised by a lack of
seasonality, which confirms the organisms’ adaptation
to disturbance events and environmental variability.
The intensively exploited continental shelf holds a
highly homogeneous benthic community dominated
by opportunistic organisms. Nevertheless, despite the
small size of the reference area in the middle of this
heavily trawled fishing ground, the functional trait
analysis detected important differences amongst the
benthic assemblages found in the fished and reference
areas.
Community homogenisation and reduction of com-
plexity can result in the loss of ecological function
(Thrush & Dayton 2002). Moreover, these changes in
benthic communities can have further consequences
for benthic invertebrates and commercial species, as
habitat structuring fauna can provide refuge and pro-
tection from predators for numerous benthic organisms
and juveniles of commercial species (Kaiser et al.
2000). We also highlight the importance of considering
‘rare organisms’ in chronically disturbed communities
to provide information about the more sensitive func-
tional components of the community. We propose a dif-
ferent approach focused on functional components
analysis, and the necessity of understanding the ecol-
ogy of disturbed communities. Managers aiming to
protect target species must consider these ecosystem
function implications.
Acknowledgements. The authors thank J. Hewitt and C.
Lundquist for their helpful comments on the manuscript. This
study was funded by the EU project RESPONSE (Q5RS-2002-
00787). We thank the participants in the Response project, as
well as all the participants in the ‘Veda’ cruises and the crew
of the RV ‘Garcia Del Cid’ for their help and enthusiasm. Data
were provided by the Fishermen’s Association of St. Carles. S.
de Juan was supported by a grant from Departament d’Uni-
versitats, Recerca i Societat of the Catalan government to
study at the National Institute of Water and Atmospheric
Research (NIWA), New Zealand. Comments from 3 anony-
mous reviewers improved the final version of the manuscript.
LITERATURE CITED
Auster P, Langton R (1999) The effects of fishing on fish habi-
tat. Am Fish Soc Symp 22:150187
Auster P, Malatesta R, Langton R, Watling L, and 5 others
(1996) The impacts of mobile fishing gear on seafloor habi-
tats in the Gulf of Maine (Northwest Atlantic): implications
for conservation of fish populations. Rev Fish Sci 4:
185202
Becker R, Chambers J, Wilks J (1988) The new S language. A
programming environment for data analysis and graphics.
Wadsworth and Brooks/Cole Advanced Books and Soft-
ware, Pacific Grove, CA
Bergman M, Hup M (1992) Direct effects of beamtrawling on
macrofauna in a sandy sediment in the southern North
Sea. ICES J Mar Sci 49:511
Blanchard F, LeLoch F, Hily C, Boucher J (2004) Fishing
effects on diversity, size and community structure of the
benthic invertebrate and fish megafauna on the Bay of
Biscay coast of France. Mar Ecol Prog Ser 280:249260
Bozzano A, Sarda F (2002) Fishery discard consumption rate
and scavenging activity in the northwestern Mediter-
ranean Sea. ICES J Mar Sci 59:1528
Bremner J, Rogers S, Frid C (2003) Assessing functional diver-
sity in marine benthic ecosystems: a comparison of
approaches. Mar Ecol Prog Ser 254:1125
Bremner J, Frid C, Rogers SI (2005) Biological traits of the
North Sea benthos: does fishing affect benthic ecosystem
function? In: Barnes PT, Thomas JP (eds) Benthic habitats
and the effects of fishing. American Fisheries Society,
Bethesda, MD, p 477489
Brown E, Finney B, Dommisse M, Hills S (2005) Effects of
commercial otter trawling on the physical environment of
the southeastern Bering Sea. Cont Shelf Res 25:12811301
Bustos-Baez S, Frid C (2003) Using indicator species to assess
the state of macrobenthic communities. Hydrobiologia
496:299309
Clarke KR, Warwick RM (1994) Changes in marine communi-
ties: an approach to statistical analysis and interpretation.
Natural Environment Research Council, Plymouth
Caddy JF (1973) Underwater observations on scallop (Pla-
copecten magellanicus) behaviour drag efficiency. J Fish
Res Bd Can 25:21232141
Collie J, Escanero G, Valentine P (1997) Effects of bottom fish-
ing on the benthic megafauna of Georges Bank. Mar Ecol
Prog Ser 155:159172
Daan N, Gislason H, Pope J, Rice J (2005) Changes in the
North Sea fish community: evidence of indirect effects of
fishing? ICES J Mar Sci 62:177188
Dayton PK, Thrush S, Agardy T, Hofman R (1995) Environ-
mental effects of marine fishing. Aquat Conserv 5:
205232
de Groot SJ (1984) The impact of bottom trawling on benthic
fauna on the North Sea. Ocean Manag 9:177190
Demestre M (1986) Les diferents comunitats naturals de la
Mediterrania. In: Diputacio de Barcelona, Lleonart J (eds)
L´Oceanografia II: Recursos pesquers de la mar catalana.
Quaderns d´ecologia aplicada 9:941
Demestre M (2006) Response of benthic communities and
sediment to different regimens of fishing disturbance in
European coastal waters (EU Project Q5RS-200200787).
Final Report. ICM-CSIC, Barcelona
128
de Juan et al.: Trawling disturbance in a benthic community
Demestre M, Sanchez P, Kaiser M (2000) The behavioural
response of benthic scavengers to otter-trawling distur-
bance in the Mediterranean. In: Kaiser MJ, de Groot SJ
(eds) Effects of fishing on non-target species and habitats.
Blackwell Science, Oxford, p 121129
Díaz J, Nelson C, Barber J, Giró S (1990) Late Pleistocene and
Holocene sedimentary facies on the Ebro continental
shelf. Mar Geol 95:333352
Fauchald K, Jumars P (1979) The diet of worms: a study of
Polychaete feeding gilds. Oceanogr Mar Biol Annu Rev
17:193284
Frid C (2003) Managing the health of the seafloor. Front Ecol
Environ 1:429436
Frid C, Harwood K, Hall S, Hall J (2000) Long-term changes
in the benthic communities on North Sea fishing grounds.
ICES J Mar Sci 57:13031309
Frid C, Robinson L, Bremner J (2005) Muddy thinking: ecosys-
tem-based management of marine benthos. In: Barnes PT,
Thomas JP (eds) Benthic habitats and the effects of fishing.
American Fisheries Society, Bethesda, MD, p 651665
Hall S (1999) The effects of fishing on marine ecosystems and
communities. Blackwell, Oxford
Hill A, Veale L, Pennington D, Whyte S, Brand A, Hartnoll R
(1999) Changes in Irish Sea benthos: possible effects of 40
years of dredging. Estuar Coast Shelf Sci 48:739750
Jennings S, Kaiser M (1998) The effects of fishing on marine
ecosystems. Adv Mar Biol 34:201352
Jennings S, Pinnegar J, Polunin N, Warr K (2001) Impacts of
trawling disturbance on the trophic structure of benthic
invertebrate communities. Mar Ecol Prog Ser 213:
127142
Kaiser M, Spencer B (1994) Fish scavenging behaviour in
recently trawled areas. Mar Ecol Prog Ser 112:4149
Kaiser M, Spence F, Hart P (2000) Fishing-gear restrictions
and conservation of benthic habitat complexity. Conserv
Biol 14:15121525
Link J, Brodziak J, Edwards S, Overholtz W, Mountain D,
Jossi J, Smith T, Fogarty M (2002) Marine ecosystem
assessment in a fisheries management context. Can J Fish
Aquat Sci 59:14291440
Lohrer AM, Thrush SF, Gibbs M (2004) Bioturbators enhance
ecosystem function through complex biogeochemical
interactions. Nature 431:10921095
McConnaughey R, Mier K, Dew C (2000) An examination of
chronic trawling effects on soft-bottom benthos of the
eastern Bering Sea. ICES J Mar Sci 57:13771388
Norse E, Watling L (1999) Impacts of mobile fishing gear: the
biodiversity perspective. In: Benaka L (ed) Fish habitat:
essential fish habitat and rehabilitation. American Fish-
eries Society, Bethesda, MD, p 3140
Palanques A, Guillen J, Puig P (2001) Impact of bottom trawl-
ing on water turbidity and muddy sediment of an unfished
continental shelf. Limnol Oceanogr 46:11001110
Pauly D, Christensen V, Dalsgaard J, Froese R, Torres F Jr.
(1998) Fishing down marine food webs. Science 279:
860863
Pearson T (2001) Functional group ecology in soft-sediment
marine benthos: the role of bioturbation. Oceanogr Mar
Biol Annu Rev 39:233267
Philippart C (1998) Long-term impact of bottom fisheries on
several by-catch species of demersal fish and benthic
invertebrates in the south-eastern North Sea. ICES J Mar
Sci 55:342352
Piet G, Jennings S (2005) Response of potential fish commu-
nity indicators to fishing. ICES J Mar Sci 62:214225
Pranovi F, Raicevich S, Libralato S, Da Ponte F, Giovanardi O
(2005) Trawl fishing disturbance and medium-term
macroinfaunal recolonization dynamics: a functional ap-
proach to the comparison between sand and mud habitats
in the Adriatic Sea (Northern Mediterranean Sea). In:
Barnes PT, Thomas JP (eds) Benthic habitats and the
effects of fishing. American Fisheries Society, Bethesda,
MD, p 545569
Ramsay K, Kaiser M, Hughes R (1998) Responses of benthic
scavengers to fishing disturbance by towed gears in differ-
ent habitats. J Exp Mar Biol Ecol 224:7389
Roth S, Wilson J (1998) Functional analysis by trophic guilds
of macrobenthic community structure in Dublin Bay, Ire-
land. J Exp Mar Biol Ecol 222:195217
Rumohr H, Kujawski T (2000) The impact of trawl fishery on
the epifauna of the southern North Sea. ICES J Mar Sci 57:
13891394
Sanchez P, Demestre M, Palanques A, Kaiser MJ (1998)
Impact of bottom trawling on the sediments and benthic
communities in the NW Mediterranean (Project no.: DG
XIV, Study 95/52. Final Report. ICM-CSIC, Barcelona
Sardá R, Pinedo S, Martin D (1999) Seasonal dynamics of
macroinfaunal key species inhabiting shallow soft-
bottoms in the Bay of Blanes (NW Mediterranean). Acta
Oecologica 20:315326
Thrush S, Hewitt J, Cummings V, Dayton P (1995) The impact
of habitat disturbance by scallop dredging on marine ben-
thic communities: What can be predicted from the results
of experiments? Mar Ecol Prog Ser 129:141150
Thrush S, Hewitt J, Cummings V, Dayton P, Cryer M, Turner
S (1998) Disturbance of the marine benthic habitat by
commercial fishing: impacts at the scale of the fishery.
Ecol Appl 8:866879
Thrush SF, Dayton PK (2002) Disturbance to marine benthic
habitats by trawling and dredging: implications for marine
biodiversity. Annu Rev Ecol Syst 33:449473
Tuck I, Hall SJ, Robertson M, Armstrong E, Basford D (1998)
Effects of physical trawling disturbance in a previously
unfished sheltered Scottish sea loch. Mar Ecol Prog Ser
162:227242
Wassenberg T, Dews G, Cook S (2002) The impact of fish
trawls on megabenthos (sponges) on the north-west shelf
of Australia. Fish Res 58:141151
Widdicombe S, Austen M, Kendall M, Warwick RM, Jones M
(2000) Bioturbation as a mechanism for setting and main-
taining levels of diversity in subtidal macrobenthic com-
munities. Hydrobiologia 440:369377
129
Editorial responsibility: Howard Browman (Associate Editor-
in-Chief), Storebø, Norway
Submitted: May 22, 2006; Accepted: August 23, 2006
Proofs received from author(s): February 28, 2007
... It is defined in a broad sense as "the collection of genome, species and ecosystems occurring in a definite region" (Boehlert, 1996). It is widely recognized nowadays that fisheries may have a severe impact on target and non-target species, lead to changes in the structure of marine habitats, influence diversity, composition, biomass and productivity of the associated biota, trigger indirect effects in marine populations and communities, and alter the structure and functioning of marine ecosystems (Boehlert, 1996;Jennings and Kaiser, 1998;Pauly et al., 2002;de Juan et al., 2007;Coll et al., 2008a;Coll et al., 2008b;Zhou et al., 2010;de Juan et al., 2020;Trindade-Santos et al., 2020). An example of this is the decline in Elasmobrach species in the Mediterranean and worldwide, and the high risk of their extinction due to fishing activities (Bradai et al., 2018;Dulvy et al., 2021;Walls and Dulvy, 2021); and similarly, the effects on benthic species and communities (de Juan et al., 2007;Coll et al., 2010;Clark et al., 2016). ...
... It is widely recognized nowadays that fisheries may have a severe impact on target and non-target species, lead to changes in the structure of marine habitats, influence diversity, composition, biomass and productivity of the associated biota, trigger indirect effects in marine populations and communities, and alter the structure and functioning of marine ecosystems (Boehlert, 1996;Jennings and Kaiser, 1998;Pauly et al., 2002;de Juan et al., 2007;Coll et al., 2008a;Coll et al., 2008b;Zhou et al., 2010;de Juan et al., 2020;Trindade-Santos et al., 2020). An example of this is the decline in Elasmobrach species in the Mediterranean and worldwide, and the high risk of their extinction due to fishing activities (Bradai et al., 2018;Dulvy et al., 2021;Walls and Dulvy, 2021); and similarly, the effects on benthic species and communities (de Juan et al., 2007;Coll et al., 2010;Clark et al., 2016). Fishing, even at sustainable levels, may cause significant biodiversity losses if compared to the status of unexploited conditions (Jennings, 2007;Collie et al., 2013 and references therein). ...
... Although the work focussed on the species diversity, a study of other properties of biodiversity and ecosystem status, related to more functional features of a community, was also undertaken. Many researchers have pointed to the importance of trophic interactions in communities and ecosystems as mediators in the complex responses of communities to biodiversity losses (Pauly et al., 2000;Worm and Duffy, 2003;de Juan et al., 2007;Coll et al., 2008b;Rochet et al., 2013;Shannon et al., 2014). Furthermore, mitigation of trawling effects on the ecosystem needs to be underpinned by detailed information on vulnerable species, and understanding vulnerability as the susceptibility of either an individual, species, population, community, habitat or ecosystem (de Juan et al., 2020). ...
Article
Full-text available
Fisheries have important impacts on marine biodiversity. In this work, combined information on the abundance, species richness, diversity indices, species composition, trophic level and vulnerability index were examined for the first-time to detect differences in five units related to trawl fishing: the fish assemblage entering the trawl codend, and the escaping, retained, discarded and landed fractions, derived by the gear and fisher selection practices. The work was based on a case study conducted in the Mediterranean Sea, using three different meshes in the trawl codend (40mm-40D and 50mm-50D diamond meshes, and 40 mm-40S square meshes) and a cover of the codends with small mesh size. In general, trawl fishing produces an escaping fraction that was always lower in abundance, richness, and vulnerability index, similar in diversity indices and trophic level, and different in species composition compared to the fish assemblage entering the codend. In almost all cases, fishers selected as landings a fraction that was the lowest in diversity indices, and the highest in trophic level. In contrast, fishers discarded a fraction that was the highest in diversity and vulnerability index, and the lowest in trophic level. Although the three codends did not differ significantly in the fraction of escapees in terms of diversity indices, trophic level, and vulnerability index, the 40S codend showed a significantly higher percentage in the escaping number of species and individuals, and less differences in the species composition; in addition, lower percentage in abundance of discards and higher of landings in the retained catch (0.6:1) than did the other two codends (0.9:1). It was suggested that an urgent modification of the trawl for the elimination of the discarded highly vulnerable species (e.g. Elasmobranchs) is needed, and that trawl species-selectivity should be improved by allowing escape or avoiding catch of the discarded fraction to minimize biodiversity losses.
... Other studies have documented the long-term functional impacts of fisheries on benthic habitats and invertebrate communities (e.g. Bremner et al., 2003;Tillin et al., 2006;De Juan et al., 2007;Lambert et al., 2014;Sciberras et al., 2018). Bottom trawling has been shown to reduce benthic production, biomass, and species richness (Collie et al., 2000;Kaiser et al., 2006;Worm et al., 2006; van Denderen et al., 2014a;Sciberras et al., 2018;Hiddink et al., 2020). ...
... Several studies have suggested that large, long-lived organisms, such as corals, sponges, and especially erect organisms taller than twenty centimeter, will disappear from areas that are regularly trawled (see SM S.6), and that the benthic community in those areas will become dominated by fast-growing, short-lived species (e.g. Bremner et al., 2003;Tillin et al., 2006;De Juan et al., 2007;Buhl-Mortensen et al., 2013Jørgensen et al., 2016). Other studies have shown that in areas with low trawling intensity, the abundance of filter-feeding organisms, attached epifauna, and larger animals was higher, whereas areas with high trawling intensity showed a higher abundance of mobile species, infaunal invertebrates, deposit feeders, and burrowing scavengers (Tillin et al., 2006;De Juan et al., 2007). ...
... Bremner et al., 2003;Tillin et al., 2006;De Juan et al., 2007;Buhl-Mortensen et al., 2013Jørgensen et al., 2016). Other studies have shown that in areas with low trawling intensity, the abundance of filter-feeding organisms, attached epifauna, and larger animals was higher, whereas areas with high trawling intensity showed a higher abundance of mobile species, infaunal invertebrates, deposit feeders, and burrowing scavengers (Tillin et al., 2006;De Juan et al., 2007). It is noteworthy from our study that some large long-lived epifaunal organisms in the benthic community survive different fishing pressures in the long term, such as the large mussels M. edulis and A. islandica, where the larger specimens of the latter reach an age of 100 years or more and were also abundant in intensively fished areas in our study. ...
Article
The short term impacts of fishing pressure were compared with the variability induced by environmental drivers on quantitative benthic community impact indicators. The different pressures were evaluated through comparative multifactor statistical analyses of their effects on macrofauna indicators in a Baltic Sea area with high natural disturbance. The area is exposed to a wide range of fishing intensities from long term non-fished areas to seasonal and annually frequently fished areas. Such evaluations are important for comparing the influence and short term variability of impact indicators from both demersal fisheries and the environment, including benthic community biodiversity (species richness), density (abundance in number of individuals), biomass, and average individual mean weight, with high spatio-temporal resolution. Environmental drivers include near bottom water current speed, salinity, temperature, and dissolved oxygen concentration, considering habitat specific and seasonal conditions. Demersal fishing-induced impacts were evident for all indicators. The highest fishing impacts were estimated in soft muddy and sandy habitats and in the second quarter of the year for all indicators, considering complex interactions. All environmental drivers, especially, current speed, had significant impacts on all indicators. The significant influences and short term variability caused by environmental drivers were of the same or larger order of magnitude as fishing impacts. Consequently, the short term influence of environmental drivers and seasonal differences in fishing pressure need to be considered when using quantitative benthic fishing impact indicators and identifying areas that are more or less resilient to fishing in relation to short- and long-term fisheries management plans.
... The polychaete Chaetozone setosa, a surface deposit feeder, was detected at all sites, but at a higher rate in deep-water ones. Generally, suspension feeders are abundant in sandy habitats and deposit feeders are predominant in muddy sediments [59], whereas carnivores are more prevalent in habitats with coarse sand [60,61]. A deposit feeder obtains nutrients by consuming organic matter attached to sediment particles and consuming detritus [62]. ...
... Thus, it was possible to predict the habitat environment simply by identifying the dominant species, which could aid aspects of marine management such as environmental impact assessments [65]. Monitoring the distribution of dominant species could allow us to detect environmental gradients across the entire habitat that occur in response to disturbances [59]. Further research is needed to evaluate the effects of environmental factors on benthic organisms in the East Sea, including the Ulleung Basin. ...
Article
Full-text available
The Ulleung Basin of the East Sea is a biological hotspot, but studies on the macrobenthos therein are lacking. To evaluate the macrobenthos species diversity and community structure in the Ulleung Basin, we conducted a survey from 2017 to 2021. A total of 262 species of macrobenthos was detected by the survey, and the average habitat density was 243 individuals/m2. The average biomass of macrobenthos was 43.7 g/m2, and the average Shannon and Wiener’s diversity index was 2.3 (0.69–3.61). The dominant species were the polychaetes Terebellides horikoshii, Chaetozone setosa, Scalibregma inflatum, and Aglaophamus sp. and the bivalvia Axinopsida subquadrata. The community structure of macrobenthos differed according to water depth, and a correlation analysis using environmental variables showed that the community was affected by water temperature, salinity, and dissolved oxygen. The macrobenthic fauna in the Ulleung Basin was greatly influenced by water depth, the sedimentary environment was different, and the dominant species were also different. In addition, there was little seasonal change compared to the East Sea coast. Our findings will facilitate further investigation of benthic ecosystems throughout the East Sea.
... small (1-3 cm) S.s medium (3.1-6 cm) S.m large (6.1-10 cm) S.l Lifespan <1 yr L.s Represents the resilience of an organism in the face of stressors (De Juan et al., 2007). Species with shorter life spans increase after disturbance due to their higher resilience capacity (Pearson and Rosenberg, 1978). ...
... As benthic taxonomic and functional patterns are known to be heavily influenced by natural gradients viz. salinity and grain size, it is challenging to trace the responses to co-existent anthropogenic stressors due to synergistic effects of multiple stressors (Egres et al., 2019;De Juan et al., 2007). Adding to the complexity, traits that are categorised as 'opportunistic' may be observed in relatively less impacted zones (Townsend et al., 1997) as observed in this study. ...
Article
Estuaries experience incessant modifications due to various stressors causing shifts in the benthic species community structure and ecosystem functioning. Two tropical estuaries along Northwest India, exposed to varied intensities of anthropogenic perturbations, were sampled seasonally for two consecutive years. Specific aims were to assess, compare and link the macroinvertebrate taxonomic and functional structures, to evaluate the resilience of estuarine benthic ecosystems by employing the multiple-trait approach and to identify major influencing environmental drivers for patterns discerned. Taxonomic and functional compositions in both the estuaries produced varied segregations along the estuarine zones, driven primarily by natural estuarine gradients like salinity and sediment grain size, despite extant anthropogenic stressors. Multiple traits contributed to the variance in benthic functioning. The Biological Trait Analysis (BTA) revealed that both the estuaries had similar trait compositions in the lower zones, while the middle and upper zones of each estuary presented different permutations of traits. The functional complexity at different estuarine sections was influenced by the variability in taxonomic composition and species dominance. However, relationships between Functional Diversity (FD) and species diversity were equivocal, signifying that taxonomic diversity may not be an efficient proxy for benthic functioning. As the zones had differential stressors and disturbance acts as a filter, discrete functional trait profiles of opportunistic traits were visualized along the potentially impacted zones. Thus, the less impacted lower zones had multiple traits, while the mid-upper zones that were subject to both anthropogenic and natural stressors had fewer traits. A more consistent functional structure, higher functional redundancy and substantial proportion of recolonisation traits (small-sized, short-lived, motile forms) suggested better resilience in one study estuary than the other one. Our study advocates that the inclusion of both taxonomical and functional metrics can provide in-depth inferences related to the macrobenthic community resilience and this coupled approach is imperative for effective future management programs.
... At the biological level, the trawl net removes animals from bottom waters, the sediment surface and shallow sub-surface. Impact studies have shown that the abundance of macrofauna and megafauna (infaunal and epifaunal) is generally reduced with corresponding changes in the community and trophic structure [6,[9][10][11][12]. Smaller body-sized fauna, however, with fast life cycles may be more resistant to trawling [8,13,14]. ...
... In addition, bioturbating organisms are particularly important as agents for irrigation and the movement of oxygen into the sediment [22,23]. With trawling impacting larger species in the sediment, continually impacted areas have a lower level of bioturbation and consequently different levels of sedimentary fluxes [8,12,13,18]. Trawling was also found to cause changes in oxygen regime, influencing the nitrogen cycle, as oxygen regulates both nitrification and denitrification [15,24]. ...
Article
Full-text available
The impact of otter trawling on the relationship between functional traits of benthic invertebrates and specific biogeochemical processes was investigated in the oligotrophic Cretan Sea. The fishery is managed through a seasonal closure during the summer. During two seasons (winter and summer) replicate samples were taken from the field from a commercial trawl ground and an adjacent control area. Environmental parameters related to sediment biogeochemistry were measured including particulate organic carbon, sedimentary organic carbon, bottom water and sedimentary chlorophyll a and phaeopigment concentrations as well as benthic oxygen consumption. A significant impact of trawling was recorded only for bottom water chlorophyll and sedimentary organic carbon. Furthermore, the links between species traits and specific ecosystem processes were affected by trawling, highlighting the importance of unique functional modalities on ecosystem functioning. The traits that mostly influenced benthic biogeochemistry in the control sites were related to bioturbation and burrowing activities. In contrast, in the trawled sites, the associated traits were related to more opportunistic lifestyles and deposit-feeding species that do not act as bioturbators. Thus, under trawling disturbance, this shift can decouple the species-sediment relations and affect nutrient cycling.
... Habitat alteration and disturbance caused by trawling has profound impacts on sessile megafauna (Jones, 1992;Williams et al., 2010;Yesson et al., 2017;Amisi et al., 2018). Although some large and mobile fauna can recolonise regions rapidly where trawling has ceased (de Juan et al., 2007;Demestre et al., 2008;Paradis et al., 2021), ecosystem recovery in soft sediments can remain limited even after 30 years (de Juan et al., 2011). Mortality due to fishing can be substantial for deep-sea benthic invertebrate megafauna (Bergman and Van Santbrink, 2000), which display negative responses to trawling due to low resilience and reduced mobility (Williams et al., 2010;de Juan et al., 2011;Ingels et al., 2014;Clark et al., 2016;Yesson et al., 2017;Amisi et al., 2018). ...
... For deep-sea organisms that are long lived and grow slowly, recolonisation and recovery can be on a timescale of centuries to millennia (Jones, 1992;Clark et al., 2016;Rijnsdorp et al., 2018;Hiddink et al., 2019). Intense fishing pressure results in a reduced abundance of macro-and megafauna and an increased abundance of burrowing epifaunal scavengers and motile burrowing infauna de Juan et al., 2007). Continued disturbance is also expected to diminish the ability of highly adapted K-selected species, such as crinoids, to establish, instead favouring more resilient, generalist, scavenging biota (Liu et al., 2011;Mangano et al., 2013). ...
Article
Full-text available
Although considerable research progress on the effects of anthropogenic disturbance in the deep sea has been made in recent years, our understanding of these impacts at community level remains limited. Here, we studied deep-sea assemblages of Sicily (Mediterranean Sea) subject to different intensities of benthic trawling using environmental DNA (eDNA) metabarcoding and taxonomic identification of meiofauna communities. Firstly, eDNA metabarcoding data did not detect trawling impacts using alpha diversity whereas meiofauna data detected a significant effect of trawling. Secondly, both eDNA and meiofauna data detected significantly different communities across distinct levels of trawling intensity when we examined beta diversity. Taxonomic assignment of the eDNA data revealed that Bryozoa was present only at untrawled sites, highlighting their vulnerability to trawling. Our results provide evidence for community-wide impacts of trawling, with different trawling intensities leading to distinct deep-sea communities. Finally, we highlight the need for further studies to unravel understudied deep-sea biodiversity.
... One drawback to life as a sedentary bottom-dweller is the reduced ability to respond effectively if one's habitat becomes inhospitable. Sedentary benthic macrofauna are disproportionately harmed by trawling (de Juan et al., 2007) or storm-induced sediment burial (Abesamis, 2018). While motile organisms are able to simply relocate to a more suitable location, sessile organisms are required to weather difficult conditions. ...
Preprint
Cassiopea sp., the Upside-Down Jellyfish, are considered to be sedentary epibenthic organisms which exhibit little to no movement on the seafloor. In this study, we use time-lapse videography of a Cassiopea population in the Florida Keys to demonstrate that Cassiopea sp. exhibit a greater degree of benthic locomotion than previously understood, with animals covering distances up to 178 cm per day. In addition, Cassiopea seem to aggregate on the bottom, with average number of neighbours consistently higher than would be expected for random distributions. Given the ability of Cassiopea to release nematocysts directly into the water column, we present this aggregation as a potential defensive behaviour in this species.
... If allowed sufficient time for recovery, the most frequently disturbed plots (F4) would likely converge with controls after roughly 28 days (as observed in the F1 treatment level). If, however, the communities were chronically disturbed and not provided sufficient time for recovery, then local community composition would be strongly influenced by disturbance frequency, as reported in other studies (de Juan et al., 2007;Sandrini-Neto et al., 2016;Thrush et al., 2003). Moderately sized, discrete, natural disturbances largely do not appear to influence successional dynamics in intertidal mudflat communities (Lindegarth et al., 2000;Norkko et al., 2010;Thrush et al., 2003). ...
Article
Full-text available
Elucidating factors (“drivers”) that influence succession after disturbance can explain ecological phenomena, including why communities vary spatiotemporally. To gain insight on drivers related to habitat availability, species availability, and species performance during succession, we conducted two field experiments on infaunal communities in intertidal mudflats, one on each of the Atlantic and Pacific coasts of Canada, that had disturbances of different type, size, and frequency. Related to habitat availability drivers, we observed that disturbance type and size, which differed between experiments, did not change end patterns of succession; however, disturbance frequency, directly assessed in one experiment, did. Dispersal of species from surrounding mudflat and water column (species availability) was the primary driver of succession, whereas local interactions between species after colonization (species performance drivers) did not have a detectable effect. We suggested that ample space and resources diffused competition and predation effects, and so species replacements did not occur in our systems, resulting in a lack of “traditional” successional dynamics as observed in other ecosystems. Our findings that community composition in intertidal mudflats is strongly influenced by species availability on two different coasts suggest that this driver may be key to variation in intertidal mudflat communities elsewhere.
... Otherwise, sedentary benthic species seem to be more resilient to the effects of trawling disturbance, e.g. starfish and crabs (de Juan et al., 2007;Mangano et al., 2013). Mobile scavengers and opportunistic taxa as the starfish A. irregularis, the decapod P. typhlops and the portunid crabs L. depurator and M. tuberculatus resulted abundant in the overall Ionian and they are generally higher in areas subjected to high levels of fishing intensity. ...
Article
Full-text available
Deep-sea communities are impacted by several anthropogenic activities, such as fisheries, which still remain one of the most damaging on the marine environments in terms of biodiversity loss and habitat degradation. The north-western Ionian Sea (Central Mediterranean) experienced long-standing trawl fishing activity with the exploitation of deep-sea demersal resources. The integrate analysis with data of both benthic, demersal and benthopelagic species collected during a time series of experimental trawl survey, yearly carried out in the Ionian basin down to 800 m in depth, allowed to asses the current status of the Ionian deep-sea faunal assemblages and their potential changes overtime. Multivariate analysis and univariate indices, modelled using Generalized Additive Model (GAM) framework, confirm a depth zonation pattern of deep-sea faunal assemblages in the study area, with the presence of two distinct epibathyal and mesobathyal groups. GAM also highlighted a temporal effect on the biodiversity indices, with significant negative trends of diversity and evenness indices as well as a significant increasing of dominance one, over the study period 2012-2020. The benthic community is characterised by more generalist species and a simplified structure, with a severe depletion in habitat-former taxa whereas the demersal and benthopelagic fauna of the Ionian Sea show a general stability in the overall structure if compared with previous studies lasting over two decades. The presence of complex and heterogenous habitats unsuitable for trawl, like cold-water coral communities and canyon systems, provide Essential Fish Habitats (EFHs) for commercial species, acting as potential renewal areas for exploited resources in the neighbouring fishing grounds. These findings encourage an ecosystem-based management including spatial considerations for the objectives of biodiversity conservation combined with those of management of fishery resources.
Thesis
Full-text available
The present PhD thesis aims to investigate the effects of the bottom trawl codend selectivity on fish populations, fisheries, and biodiversity. At the same time, it is investigating the best scenario for the codend, in line with the current EU Mediterranean legislation, by testing different types of meshes in the codend. Finally, it is also seeking to identify the main behaviour patterns of fish that can support the improvement of the trawl codend selectivity.
Article
Full-text available
Summary 1. The behaviour of scavengers and predators was studied in response to otter-trawling disturbance in muddy sediments in the north-west Mediterranean. 2. Repeated trawling with a commercial fishing gear over the same plotted coordinates depleted the abundance of commercial fishing importance species such as hake. However, smaller scavenging and predatory species increased in abundance significantly with time. 3. As in previous studies, the aggregative response of scavengers was short-lived and lasted no more than several days, which indicate that additional food resources made available by the trawling activities were rapidly consumed.
Chapter
Full-text available
Demersal fishing gear affects seabed habitats both directly and indirectly and modifies the processes and dynamics of benthic communities. At present, scientific attention is focused on using the functional approach to better understand the constraints that drive the recolonization of benthic fauna subjected to fishing disturbance. The Northern Adriatic Sea has an extensive trawlable area that is intensively exploited by a variety of trawling gear. Among these, the rapido, a beam trawl used to catch flatfishes on mud substrate and pectinids on sand, appears to have the greatest impact on the benthic habitat. We used various functional indicators (diversity indices, community structure, trophic groups, production analysis, and exergy) to compare the medium-term (9-month) macroinfaunal recolonization processes in sand and mud habitats treated with a single experimental rapido haul. Recolonization was found to be a community-wide process. The early successional stages in the sand habitat were dominated by scavenging organisms that peaked in abundance 7 d after the treatment. In the mud habitat, the trend in scavenger activity was less distinct, although a peak in abundance was recorded 1 month after the treatment. The functional analyses revealed that complete recovery required at least 9 months in both habitats. Finally, data collected on the fishing ground near the sand experimental area were analyzed in order to investigate the chronic disturbance caused by commercial trawling. The fishing ground samples showed a higher spatial heterogeneity than the sand experimental samples. Total abundance, total biomass, and production values in the fishing ground were comparable with the lowest recorded values in the sand experimental area, and the exergy differences suggested that the fishing ground’s benthic community remained in an early successional stage.
Article
Full-text available
We examined a suite of abiotic, biotic, and human metrics for the northeast U.S. continental shelf ecosystem at the aggregate, community, and system level (>30 different metrics) over three decades. Our primary goals were to describe ecosystem status, to improve understanding of the relationships between key ecosystem processes, and to evaluate potential reference points for ecosystem-based fisheries management (EBFM). To this end, empirical indicators of ecosystem status were examined and standard multivariate statistical methods were applied to describe changes in the system. We found that (i) a suite of metrics is required to accurately characterize ecosystem status and, conversely, that focusing on a few metrics may be misleading; (ii) assessment of ecosystem status is feasible for marine ecosystems; (iii) multivariate points of reference can be determined for EBFM; and (iv) the concept of reference directions could provide an ecosystem level analog to single-species reference points.Nous avons étudié une série de >30 métriques abiotiques, biotiques et humaines de l'écosystème de la plate-forme continentale du nord-est des É.-U. à l'échelle de l'association, de la communauté et du système couvrant plus d'une trentaine d'années. Nos objectifs principaux étaient de décrire l'état de l'écosystème, d'améliorer la compréhension des relations entre les processus principaux de l'écosystème et d'évaluer des valeurs de référence potentielles à utiliser dans une approche écosystémique de gestion de la pêche (EBFM, « ecosystem-based fisheries management »). Pour ce faire, nous avons examiné les indicateurs empiriques de l'état de l'écosystème et nous avons utilisé les méthodes multidimensionnelles courantes pour décrire les changements dans le système. Nous avons trouvé que (i) il faut une série de métriques pour caractériser de façon précise l'état de l'écosystème et, inversement, la concentration sur un petit nombre de métriques peut induire en erreur; (ii) il est possible d'évaluer l'état des écosystèmes marins; (iii) on peut déterminer des valeurs de références multidimensionnelles pour usage dans l'EBFM; et (iv) le concept de direction de repérage peut fournir à l'échelle de l'écosystème un analogue aux valeurs de référence utilisées dans le cas d'une seule espèce.[Traduit par la Rédaction]
Article
Full-text available
The diets of gurnards Aspitrigla cuculus and Eutrigla gurnardus, lesser-spotted dogfish Scyliorhinus canicula and whiting Merlangius merlangus were examined to determine whether they migrated into recently trawled areas to feed on animals that may be damaged or dislodged by the action of a 4 m beam trawl. Gurnards and whiting increased their intake of prey after an area had been fished. In particular, they increased the proportion of the amphipod Ampelisca spinipes in their diets. Beam trawling damaged the purple burrowing heart urchin Spatanguspurpureus, scallop Aequipecten operculans, Ensisspp. and Laevocardium sp., exposing internal tissues which were then eaten by whit- ing. Some mobile invertebrate scavengers, such as Pandalusspp., only occurred in diets after the area had been fished, suggesting that these animals were also scavenging over the trawl tracks. Observa- tions of the seabed using a side-scan sonar revealed a greater concentration of fish marks around the trawl tracks than in adjacent unfished areas. Our results indicate that fish rapidly migrate into beam trawled areas to feed on benthic animals whlch have been either damaged or disturbed by fishing or on scavenging invertebrates. In areas where certain benthic communities occur, beam trawling inten- sity may be such that it creates a significant food resource for opportunistic fish species. This is a possi- ble mechanism whereby long-term community structure could be altered by fishing activity.
Article
Full-text available
Current information on the eating habits of polychaetes are reviewed. The polychaete families are arranged in alphabetical order and the feeding habits of each are summarized. Feeding guilds are defined based on a joint consideration of food, eating habits, and locomotory patterns. An understanding of the sympatric occurrance in the deep sea of several congeners with extremely limited morphological differentiation was of particular interest. This review contains 355 references, 23 figures, and 33 tables.
Article
Functional analysis of the structure of marine benthic communities has a chequered history. On the one hand, proponents of such analyses point to the increased understanding of the ecology of communities gained when their component species are grouped into units with comparable functions. On the other hand, critics of the concept suggest that such analyses are both superficial and misleading without a much better knowledge of the life histories and behaviour of benthic organisms than is available at present. Both these points of view are considered in the light of experience gained from some recent functional group analyses of soft-sediment benthos. It is concluded that these analyses have provided a clearer understanding of the influence on benthic faunal distributions of changing environmental conditions along vertical and latitudinal gradients. They have also produced some new insights into the relative importance of differing environmental variables in structuring benthic communities. The role of bioturbation in soft-sediment benthic communities is examined in some detail. The various ways in which bioturbatory roles have been partitioned into functional groups are discussed and a tentative scheme for characterising such roles on the basis of behaviour is suggested as an adjunct to process-based partitioning. It is suggested that the delimitation of groups based on a number of functions that are adapted to a particular environmental pressure is the best means of gaining ecological insights from this type of analysis.
Article
Late Pleistocene—Holocene history of the Ebro continental shelf of northeastern Spain is recorded in two main sedimentary units: (1) a lower, transgressive unit that covers the shelf and is exposed on the outer shelf south of 40°40′N, and (2) an upper, progradational, prodeltaic unit that borders the Ebro Delta and extends southward along the inner shelf. The lower transgressive unit includes a large linear shoal found at a water depth of 90 m and hardground mounds at water depths of 70–80 m. Some patches of earlier Pleistocene prodelta mud remain also, exposed or covered by a thin veneer of transgressive sand on the northern outer shelf. This relict sand sheet is 2–3 m thick and contains 9000–12,500 yr old oyster and other shells at water depths of 78–88 m.
Article
The North Sea has been subjected to fishing activity for many centuries. However, improvements in both fishing vessels and trawling gears since the early 1900s have meant that fishing intensity has increased. A resultant increase in the areas trawled and the use of heavier and potentially more destructive gears probably had effects on the marine community. Information on benthic communities within the North Sea, from both published and unpublished sources, has been compiled to provide a long-term data set of changes in the marine benthos on five selected fishing grounds over 60 years. In two of these (Dogger Bank and Inner Shoal), there was no significant difference in community composition between the early 1920s and late 1980s. In the remaining three areas (Dowsing Shoal, Great Silver Pit, and Fisher Bank) significant differences were observed. However, these were the result of changes in abundance of many taxa rather than large-scale losses of sensitive organisms. These results suggest that fishing has influenced benthic communities in the North Sea. The possibility remains that fishing-induced changes had occurred at the Dogger Bank and Inner Shoal prior to the 1920s.