Relationship between razor clam fishing intensity and potential changes in associated benthic communities.
KEY WORDS: macrobenthic communities, bivalve dredging, natural disturbance, functional traits, razor clam, Ensis
Concern about the environmental effects associated with bivalve fishing has recently increased, as indicated by the growing number of studies and reviews undertaken worldwide (Eleftheriou & Robertson 1992, Hall 1994, Thrush et al. 1995, Kaiser & Spencer 1996, Jennings & Kaiser 1998, Turner et al. 1999). It is known that dredging can have an immediate effect on benthic fauna and can also change the physical environment of marine ecosystems (Eleftheriou & Robertson 1992, Thrush et al. 1995, Kaiser et al. 1996, Gilkinson et al. 2003, Gilkinson et al. 2005, Kaiser et al. 2006). The degree of the impacts can range from minor (Jenkins et al. 2001) to severe (Koslow et al. 2001), and strongly depends on the gear used, water depth, nature of the substrate, benthic communities being impacted, and fishing intensity (Bergman & Hup 1992, Kaiser et al. 1996, Jennings & Kaiser 1998, Tuck et al. 1998, Tuck et al. 2000, Gaspar et al. 2003a, Kaiser et al. 2006, Gaspar & Chicharo 2007).
To date, several field experiments have focused on immediate and short-term effects of different types of dredges on benthic communities (Hall et al. 1990, Kaiser & Spencer 1996, Kaiser et al. 1996, Bergman et al. 1998, Tuck et al. 2000, Chicharo et al. 2002a, Gaspar et al. 2003a, Gilkinson et al. 2003, Hauton et al. 2003, Pranovi et al. 2004, Gilkinson et al. 2005, Constantino et al. 2009), as well as their environmental effects in numerous habitats (Kaiser et al. 1996, Kaiser et al. 2006). However, the majority of these studies were mostly focused on small-scale changes of community patterns resulting from a single dredging event, rather than on the impacts resulting from ongoing dredging pressure (Lindegarth et al. 2000a, Bradshaw et al. 2002). It is widely accepted that growing fishing pressure, along with the consequent intensive and repeated dredging in the same area, may lead to longterm changes in benthic assemblages (Kaiser et al. 1996, Kaiser et al. 2005). Studies on the long-term effects of fishing are relatively scarce, probably resulting from the temporal and spatial natural variability of marine benthic communities and the lack of suitable undisturbed areas (Thrush et al. 1995, Frid et al. 1999, Veale et al. 2000, de Juan et al. 2007). The paucity of nonfished grounds constitutes a major constraint for detecting long-term impacts of bivalve dredging in benthic communities (Tuck et al. 1998, Lindegarth et al. 2000a).
In previous long-term studies, it has been reported that large-bodied and long-lived benthic fauna are more vulnerable to removal or damage by the fishing gear (Jennings et al. 2001a, Kaiser et al. 2005), whereas highly mobile, well-armored (exoskeletons or thick shells), and scavenging species tend to be more resilient to fishing effects, becoming the dominant fauna in areas subjected to intense disturbance (Ramsay et al. 1997, Kaiser et al. 2000, Schratzberger et al. 2002, Garcia et al. 2006, Morello et al. 2006). Therefore, according to their functional traits, organisms can be differentially affected by bivalve fishing. An analysis of biological traits provides a more in-depth knowledge of the ecological functions in a marine benthic community than the relative taxon composition and the trophic groups approach (Bremner et al. 2003). Functional diversity incorporates the interactions between organisms and their environment, and, in opposition to the taxon composition, which is geographically dependent, provides a more robust method to be applied on an international scale (Bremner et al. (2003) and references therein).
The exploitation of subtidal bivalve beds along the Portuguese coast started during the late 1960s (Gaspar et al. 2003a), and studies assessing immediate and short-term effects of Portuguese clam dredging have already been carried out (Chicharo et al. 2002a, Chicharo et al. 2002b, Alves et al. 2003, Gaspar et al. 2003a). However, long-term effects of different fishing intensity on macrobenthic communities remain unknown, and are a handicap for future fisheries management. Fisheries management needs to address the sustainability of commercial stocks and at the same time aim to minimize the direct and indirect impacts of fishing on the other components of the ecosystem. Currently, there are a number of management options that aim to reduce marine biodiversity threats--namely, the creation of marine protected areas, spatial and temporal management of seafloor areas, and gear limitation or modification.
The current study tests whether the recovery period of macrobenthic communities depends on fishing intensity aiming to achieve the following objectives: (1) to assess the relationship between fishing intensity and changes in macrobenthic community patterns, concerning medium-term impacts and potential recovery; (2) to examine whether main taxonomic groups can be differentially affected by bivalve dredging; and (3) to evaluate changes in biological traits in relation to fishing intensity. The experiments were undertaken in an unexploited area divided into 3 areas: (1) an unfished area (control), (2) a moderately fished area (fishing for 3 h in 2 consecutive days), and (3) a heavily fished area (fishing for 7 h in 4 consecutive days). With this experimental strategy, we aimed to avoid differences among areas resulting from geographical and environmental physical conditions.
MATERIAL AND METHODS
The study was undertaken in an unexploited area off Lagos (southern Portugal) (Fig. 1). This area had been an Ensis siliqua ground, which became overexploited during the early 1990s. Since 1996, no bivalve dredging has occurred in this area (Gaspar et al. 2009). The oceanographic settings for the Algarve coast are given in detail in Constantino et al. (2009). Offshore wave data for the survey period was obtained from the Portuguese Hydrographical Institute. During the experimental period, mainly low-energy conditions (Hs < 1 m) were observed, except for a single period between 30 days and 120 days after experimental fishing when significant wave height attained a maximum of 4 m.
Prior to the selection of sampling areas, sediment samples were collected in the Bay of Lagos. It is known that benthic communities naturally vary by type of sediment. Therefore, to ensure that potential differences in community structure between nonimpacted and impacted areas were the result of dredging, these areas should exhibit similar sediment characteristics. With this purpose, 18 sediment samples for granulometry and total organic content were collected at a depth between 8 m and 10 m with a Van Veen grab. Grain-size analysis was performed by wet separation of coarse (sand and gravel) and fine (silt and clay) fractions using a 63-gin sieve. Sediment granulometry was determined by sieving and pipette analyses (Suguio 1973). Grain-size determination was made according to phi scale intervals after Wentworth (1922), and type of sediment in terms of its dominant grain-size components (gravel, sand, mud) followed the classification of Folk (1954). Granulometric parameters of sediments were established using the GRADISTAT method (Blott & Pye 2001) to obtain the mean and sorting coefficient according to the classification established by Folk and Ward (1957). To quantify organic matter in the sediment, samples were oven dried at 60[degrees]C for 24 h, and ash-free dry weight was determined after combustion in a muffle furnace at 450[degrees]C for 24 h.
[FIGURE 1 OMITTED]
Sampling of Macrobenthic Communities
Fishing operations were carried out onboard the IPIMAR's research vessel Diplodus using the Portuguese small razor clam dredge (fishing gear design can be found in Gaspar et al. (1999)). This fishing gear weighs around 80 kg and is comprised of a metallic frame, a toothed lower bar at the dredge mouth (64 cm in length with a tooth length of 30 cm, spaced 2.5 cm apart), and a net bag with a mesh size of 3 cm (stretched mesh) to retain the catch. The dredges used in the current study were similar to the ones used by professional fishermen.
Sampling followed a BACI (before-after, control-impact) experimental design to investigate the potential effects of bivalve fishing in 2 experimental treatments by comparison with a control area. The experimental design included 3 areas of approximately 2,000 [m.sup.2] each: one control area where no fishery was carried out, one moderately impacted area (MI), and one highly impacted area (HI), with different levels of fishing intensity. In the MI area, fishing operations were undertaken during 2 consecutive days for 3 h, whereas in the HI area, fishing operations were carried out for 4 consecutive days for 7 h. To minimize interactions between areas, each area was located about 500 m away from the other. The location of each area was marked with a buoy, and geographical coordinates were recorded by DGPS. Fishing operations were carried out in circles around the buoy. During experimental dredging, the vessel position was recorded every 15 sec using the onboard DGPS. This procedure allowed us to estimate the total area dredged in each experimental area. The length dredged between 2 consecutive positions was estimated using the following expression:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
where lat1 is the latitude value at point 1, lat2 is the latitude value at point 2, lon1 is the longitude value at point 1, and lon2 is the longitude value at point 2.
The estimated total dredged area in the MI and HI areas was 9,954 [m.sup.2] and 22,907 [m.sup.2], respectively. Sampling occurred from July 1 until October 28, 2007, in a sandy bottom at a 9-m depth. Because of logistical restrictions imposed by (1) the inexistence of multiple nonfished areas along the southern Portuguese coast; (2) the need to carry out the experiment in areas in the same depth range, because it has been shown that fishing impacts on macrobenthic communities in this coastal area are depth dependent (Constantino et al. 2009); and (3) the intention to simulate real fishing operations, the manipulative experiment undertaken in the current study was unavoidably pseudo-replicated. To minimize the problem of pseudo-replication, in each treatment area, 3 sites (separated by approximately 20 m) were randomly collected to have a measure of small-scale variability in each area. In each site, 3 macrobenthic replicates, 1 m apart, were collected by divers using plastic core tubes (9 cm in diameter, 20 cm deep, 0.006 [m.sup.2]). To increase the sampled area, each replicate consisted of 3 core samples (total replicate area, 0.018[m.sup.2]). Samples were collected before fishing operations, immediately after fishing, and 1, 2, 3, 7, 14, 30, and 120 days after the first fishing event. Although fishing operations continued for 2 days (MI) and 4 days (HI), for the sake of simplicity, in this article, the expression "after fishing" refers to the first fishing event. The macrofauna samples were sieved in situ using a 0.5-mm mesh sieve and were preserved in a 4% solution of buffered formalin. In the laboratory, animals were hand sorted into major taxonomic groups, identified to the lowest practical taxonomic level, and counted. Biomass, expressed as ash-free dry weight ([+ o r-]0.0001 g), was obtained by loss on ignition (450[degrees]C for 3 h) of oven-dried organisms (70[degrees]C for 48 h) and determined per taxa, replicate, and area.
For descriptive statistics, mean and SD were calculated for abundance and number of species, considering the entire community and the most abundant taxonomic groups separately (Amphipoda, Polychaeta, Gastropoda, and Bivalvia). Average taxonomic distinctness ([[DELTA].sup.+]) was calculated considering 6 taxonomic levels (species, genus, family, order, class, and phylum) (Warwick & Clarke 1995). To determine whether different patterns can be obtained using different major taxonomic groups independently, this index was calculated for all taxa combined, and for Crustacea, Annelida, and Mollusca, as suggested by Ellingsen et al. (2005). Macrobenthic community structure was also analyzed in terms of diversity (Shannon-Wiener index H', [log.sub.10]), Margalefs species richness (d), Pielou's equitability (J'), and biomass per area and per sampling period. Abundance of polychaete families during the study period was also assessed because this group was one of the main taxonomic groups found, and several polychaete species have already been shown to be indicative of environmental impacts (Levin 2000, Bustos-Baez & Frid 2003, Cardoso et al. 2007).
For the feeding functional group analysis, the identified taxa were assigned to at least one of the following feeding functional modes: suspension feeding, deposit feeding, carnivore, and herbivore (adapted from Sprung (1994), Gaston et al. (1998), Mancinelli et al. (1998), Roth and Wilson (1998), and Gaudencio and Cabral (2007)). After fishing, an increase of carnivorous feeding and a decrease in suspension feeders is commonly observed (Lindeboom & de Groot 1998, Kaiser & Spencer 1994, Frid et al. 2000, Veale et al. 2000); therefore, the ratio abundance of carnivores to suspension feeders was analyzed for all areas during the study period.
The functional trait analysis (de Juan et al. 2007) considered 4 biological traits reflecting some aspects of species morphology and behavior: animal size (<1 cm, 1-5 cm), body design (vermiform, scale or chitinous plates, shell), presence or absence of an external protective structure (e.g., tube or shell in the case of pagurids that are not intrinsic to the animal), and adult mobility (sedentary, low mobility, and medium mobility). The 10 most abundant taxa per area and per sampling period were pooled, resulting in a matrix of 35 taxa, and were classified based on literature and expert knowledge. This selection aimed to focus the analysis on the taxa with the highest contribution to the differences among treatments during the study period. For the species without any information, we looked for similar species.
The linear model used for the uni- and multivariate hypotheses being tested contained 3 factors: time (before, immediately after, and 1, 2, 3, 7, 14, 30, and 120 days after fishing; number of levels, a = 9; fixed), treatment (control, MI, and HI areas; number of levels, b = 3; fixed), and site (number of levels, c = 3; random, nested to treatment). A 3-way nested ANOVA was used because it provides an independent estimate of the variability among replicates and, consequently, a test of small-scale variability (Underwood 1997). The current design allows us to separate temporary and persistent effects of fishing: The existence of a significant effect of treatment indicates a persistent effect, as differences between treatments do not change throughout the experiment); a significant interaction time x treatment indicates that differences among treatments were not consistent during the experimental period (Ragnarsson & Lindegarth 2009). These are the main terms to assess for fishing impacts, whereas site (treatment) and time x site (treatment) give measures of small-scale variability and its consistency over time, respectively. Time effects are not discussed because they are not related to the hypotheses being tested. To maximize the statistical power of univariate analyses, postpooling and elimination of sources of variance [site (treatment) and time x site (treatment)] were applied whenever appropriate (P > 0.25) (Underwood 1997). Post hoc multiple comparisons were performed using the Student-Newman-Keuls (SNK) test to evaluate at which sampling occasion sampling areas were different. For the functional trait data, a 3-way nested multivariate analysis of variance (MANOVA) was undertaken for each trait. When a significant difference was detected, ANOVAs were performed for each one of the categories of the biological trait being tested, and post hoc multiple comparisons were applied whenever appropriate using the SNK test (e.g., Lenihan et al. 2003). Prior to analysis of variance, data were analyzed to test for normality (Kolmogorov-Smirnov test) and homogeneity of variance (Cochran's test). Heterogeneity was removed by log(x + 1) or [(x + 1).sup.0.5] transformations.
To identify similarities and dissimilarities between sample groupings, to assess potential fishing impacts on macrobenthic community structure, nonmetric multidimensional scaling (NMDS) ordinations using the Bray-Curtis similarity index (group-average linkage method) were calculated from square root-transformed abundance data. ANOVA, MANOYA, and normality/homogeneity of variance tests were performed using the statistical packages STATISTICA 6 and WinGMAV5 (EICC, University of Sydney). Multivariate (NMDS) and univariate methods were applied using the PRIMER v5.0 software package (Clarke & Warwick 1994) and MATLAB (Mathworks Inc.).
All samples were characterized by the predominance of the sand fraction (ranging from 85.8-99.9% by weight); the silt/clay fraction was always less than 0.01%. According to the phi values determined (range, -0.14-2.59) sediment in all samples but one were classified as fine-sand sediment. The exception occurred in the sample collected at a 10-m depth in the western part of Lagos, which was comprised of 14% gravel and therefore was classified as very coarse sand. In general, the organic content of the sediment was extremely low (<0.5%). Based on these results, the selected experimental areas consisted of homogeneous fine-sand sediment.
A total of 23,763 individuals distributed among 265 taxa were recorded. Amphipoda dominated both in terms of abundance and number of species followed by Polychaeta. Gastropoda and Bivalvia were also well represented, with Bivalvia presenting a greater number of taxa and Gastropoda presenting greater abundance. During the study period, the five most abundant taxa were the polychaete Spio cf. armata and the amphipods Photis spp., Cheirocratus sundevalli, Aora typica, and Phtisica marina. The taxonomic groups with a greater contribution for total biomass were Bivalvia, followed by Gastropoda, Polychaeta, and Echinodermata. In each group, the ranked species that contributed most to biomass were Thracia papyracea, Mesalia mesal, Hyalinoecia bilineata, and Echinocardium cf. cordatum.
The overall patterns of abundance and number of taxa for all community data and main taxonomic groups evidenced a general decrease in these variables (except for bivalves) after fishing in impacted areas compared with the control, but, after 3 days there was a clear, increasing trend (Fig. 2). Accordingly to these observations, the 3-way nested ANOVA indicated a significant interactive effect--time x treatment--for all variables except for gastropods (Table 1). The SNK test confirmed the temporary negative effect of fishing in impacted areas for abundance and number of taxa (Fig. 2). This negative effect was observed between fishing immediately after fishing (amphipod and polychaete abundance) and 7 days after fishing (abundance of Polychaeta; Fig. 2). The polychaete assemblage seemed to be more affected with increasing fishing intensities because differences between the control and HI areas were more long-lasting (Fig. 2). For bivalves, the pattern was not as evident (Fig. 2). Abundance and number of gastropod taxa showed contrasting results. Although no significant effects were detected in the mean number of taxa, abundance was significantly and persistently lower in HI areas compared with MI and control areas (Fig. 2). A significant effect on small-scale variability was detected for the number of amphipod taxa (Table 1), indicating that differences among sites in treatments could be the result of factors other than fishing or that responses of these taxa were not consistent among sites.
Negative temporary effects on Shannon diversity (H'), Margalef species richness (d), and biomass were also detected (Table 1). For these variables, a decrease after fishing in impacted areas compared with the control area was perceptible, in accordance with the results previously described for abundance and number of species (Fig. 3). This negative effect after fishing was supported by SNK tests, with significant differences between impact and control areas persisting until 2 days after fishing (Fig. 3). Moreover, differences between the two impacted areas were also detected 2 days after fishing (Fig. 3). During the study period, the evenness index presented very small oscillations, but small-scale variability was found to be significant (Fig. 3, Table 1) and, therefore, fishing impacts based on this variable cannot be drawn.
Regarding the average taxonomic distinctness ([[DELTA].sup.+]; Fig. 4), in general, molluscs presented higher taxonomic distinctness than the 2 other dominant groups (Polychaeta and Crustacea). For all taxa, only subtle changes were observed during the experimental period between impacted and control areas (Fig. 4). The mean values of the average taxonomic distinctness of Crustacea and Polychaeta were smaller in impacted areas (especially HI) after fishing; but after 3 days, values increased and became comparable with the control area (Fig. 4). ANOVA showed that only for Crustacea was temporary fishing effects discernible (Table 1). The SNK test indicated that effects were noticed 1 day (Control = HI > MI; P < 0.05) and 2 days (Control = MI > HI; P < 0.05) after fishing. Despite the highest differences between control and impacted areas observed for molluscs (Fig. 4), no significant effects were detected, which could be related to the high variability between samples that could hinder the detection of statistically significant differences.
Behavior of Polychaete Families
The analysis of polychaete family abundance during the study period showed that the families Pectinariidae and Sabellidae (mainly comprising tube building, sedentary, and suspension-feeding species) were significantly, and persistently, affected by fishing (Table 1). The lack of significant interaction--time x treatment--indicated that the effect persisted throughout the experimental period. Nevertheless, Sabellidae seemed to be less affected than Pectinariidae because significant lower mean abundance values were only detected in HI area compared with the control area. For Pectinariidae, differences were also detected for the MI area, but no differences were detected between the 2 fished areas. Temporary effects of fishing were detected for the family Spionidae (mainly comprising tube building, deposit-feeding species; Table 1). Post hoc multiple comparisons indicated that significantly lower mean values were registered in fished areas compared with the control area 1-7 days after fishing, and that between 2 days and 7 days, the MI area also presented higher values than the heaviest impacted area (Fig. 5). During subsequent sampling occasions, abundance increased considerably in impacted areas, and abundance become significantly higher in fished areas than in the control area 30 days after fishing (Fig. 5). Capitellids (tube building, deposit-feeding species) showed a peculiar pattern because they were almost nonexistent in all sampled areas until 7-14 days after fishing, when a peak (although not significant) was observed in the MI and HI areas. Abundance decreased thereafter until the end of the study period (Fig. 5). Nevertheless, as a result of the high variability, no significant differences between treatments were detected (Table 1). For Goniadidae, a significant effect on small-scale variability was detected, which hampered any further inspection of fishing effects (Table 1). On the other hand, Terebellidae, Opheliidae, and Hesionidae presented significant differences between control and impacted areas before fishing (Fig. 5). Therefore, differences detected afterward cannot be properly assigned to fishing because they could be related to natural variability.
[FIGURE 2 OMITTED]
The analysis of the macrobenthic community trophic structure (Fig. 6A) did not show differences in the dominance of functional groups with increasing fishing intensity, because control and impacted areas presented similar percentages of each group throughout the study period. Deposit feeders dominated in all areas, which might hamper the detection of changes concerning other feeding modes.
A significant, although not persistent, effect of fishing was also detected for the ratio of carnivorous feeders to suspension feeders (Table 1). The SNK test showed that immediately after (HI) and 3 days after (MI) fishing, impacted areas presented higher mean ratios compared with the control area (Fig. 6B). On the other hand, significantly higher ratios were also detected 120 days after fishing in the control area compared with fished areas, but these differences cannot be attributed to fishing (Fig. 6B).
Functional Trait Analysis
Concerning functional traits, the impacted areas generally presented a decrease of all analyzed traits after fishing (Fig. 7). Tests of the hypothesis that fishing affected the spatial and temporal variability of biological traits (i.e., site (treatment) and time x site (treatment)) showed that the estimated effects of both terms were negligible (P > 0.25). Therefore, post hoc pooling of sites and the elimination of the time x site (treatment) term was performed. Once more, fishing impacts were detected for biological traits because significant interactive effects--time x treatment--were detected for all categories being tested. SNK results for the interaction time x treatment for each trait are presented in Table 2.
Concerning the existence (or not) of external protection, impacted areas showed a significant decrease of both tubiculous and nontubiculous specimens compared with the control 1 and 3 days after fishing (Table 2). In all study areas, a clear increase of all biological traits was observed 3 days after fishing, with abundances usually higher than observed before fishing (Fig. 7). A similar pattern was detected for body size (Fig. 7). With regard to body design and mobility, not only a significant decrease of biological traits considered was detected between control and impacted areas until 3 days after fishing at most, but an effect of fishing intensity also become apparent. Indeed, for some of the categories (scales or chitinous plates, vermiform, low mobility) 2 days after fishing, means were also significantly lower in HI areas than in MI areas (Table 2).
[FIGURE 3 OMITTED]
Nonmetric multidimensional scaling ordination plots (NMDS) for all taxa combined and for the most representative taxonomic groups are shown in Figure 8. This analysis showed that macrobenthic communities from the 3 selected areas were very similar at the beginning of the experiment. In general (except for gastropods), all samples collected 120 days after fishing were clearly separated in the NMDS plots. For all taxa, amphipod and polychaete samples collected 1 day and 2 days after fishing in MI and HI areas, respectively, were also clearly separated in the plots (Fig. 8). Samples collected in the impacted areas immediately, 1 day (HI), 2 days (MI), and 3 days after fishing were in an intermediate position between the 2 previously mentioned and the remaining samples from the control area and the impacted areas that were clustered together (Fig. 8).
Fishing Impacts and General Community Patterns
Macrobenthic communities are largely controlled by the physical environment and, consequently, variation in environmental conditions may lead to changes in species distribution and abundance (Frid et al. 2000). Several authors pointed out that the degree to which the benthic community is affected and the subsequent recovery time is largely related and proportional to the intensity of fishing activities (Collie et al. 2000, Kaiser et al. 2000). In the current study, for most of the community variables, temporary fishing effects were detected. These effects were mainly observed between 1 day and 3 days after fishing, and resulted from a significant decrease in means in impacted areas compared with the control area. More rarely, effects were detected immediately after fishing or persisted beyond 3 days. The existence of small-scale changes in benthos after fishing procedures--namely, the reduction in abundance, number of taxa, diversity, and biomass--are well documented (Dayton et al. 1995, Thrush et al. 1995, Kaiser & Spencer 1996, Alves et al. 2003, Kaiser et al. 2006, Ragnarsson & Lindegarth 2009). These changes reflect the direct and indirect impacts of the fishing gear on macrobenthic communities through: (1) the removal of organisms from the sediment together with the target taxa, (2) displacement to other areas, and (3) damage of the specimens or changes in their position in the sediment resulting from sediment disturbance, with consequent increased vulnerability to predators (see, for example, Kaiser and Spencer (1996), Veale et al. (2000), and Jennings et al. (2001a)). However, with regard to gastropod abundance, a persistent effect was detected, indicating that the mean abundance of this faunal group was systematically lower in impacted areas than in the control area throughout the experiment. Different patterns between taxonomic groups also emerged for average taxonomic distinctness. Only for crustacean data was a temporary effect of fishing detected, with significantly lower taxonomic diversity being observed 1-2 days after fishing in impacted areas. Ellingsen et al. (2005) had already reported different patterns on taxonomic diversity depending on the way data are analyzed. Different conclusions can be obtained when the entire community is analyzed or when only a taxonomic group is used as bioindicator. Therefore, the independent analysis of the main taxonomic groups should be considered to increase clearness in benthic patterns.
[FIGURE 4 OMITTED]
The feeding guild approach offers the possibility of identifying the relationship between trophic structure and ecosystem changes because physical disturbance may alter the dominant food source with consequences for existing fauna. There is extensive evidence that long-term changes induced by commercial fisheries may become apparent through shifts in the dominance of feeding modes (Lindeboom & de Groot 1998, Jennings et al. 2001b). The dead and injured fauna left on the seafloor or exposed in the dredged tracks attracts mobile scavengers and predators known as opportunistic species (Kaiser & Spencer 1994, Frid et al. 2000, Veale et al. 2000, Gaspar et al. 2003b). Subsequently, the persistent disturbance of the sea bottom may benefit these opportunistic organisms, increasing their abundance (Frid et al. 2000). By contrast, other functional groups, such as suspension-feeding species, would suffer a decrease in intensely dredged areas (Lindeboom & de Groot 1998). Despite several studies that have documented an increase in scavenger and predator species in areas under intense dredging activity (Ramsay et al. 1997, Rumohr & Kujawski 2000), the current study did not detect any clear shift in the macrobenthic feeding guild composition, because the control area and the 2 impacted areas presented similar percentages of different feeding modes, with a clear dominance of deposit feeding. Similar findings were also observed by Jennings et al. (2001b), indicating the continuous production of species that resist the effects of trawling. Bremner et al. (2003) showed that feeding mechanisms were important distinguishing features between communities, but less relevant than mode of attachment, body form, and mobility. The relative stability of the deposit-feeding mode before fishing probably dictated the absence of changes in the trophic structure because, once established, the detritus-feeding group can hamper the recovery of suspension feeders either by consuming or smothering the potential recruits (Dayton et al. 1995). The frequent natural disturbance in the study area with resuspension of sediment may have a similar effect to fishing, with deleterious impacts for suspension feeders. Nevertheless, a significant increase in the ratio of carnivorous feeding to suspension feeding was detected in MI and HI areas a few days after the beginning of fishing operations. This increase may reflect a temporary aggregation of scavengers in the impacted areas, a decrease of species with suspension feeding mode due to fishing or both. A detailed analysis of polychaetes families showed that families Pectinariidae and Sabellidae, mainly comprising species with suspension feeding modes (besides being sedentary) were persistently affected by fishing, which could at least partially support the result found for that ratio.
[FIGURE 5 OMITTED]
Functional Traits Analysis
Several trawling impact studies demonstrated that organism response to disturbance and consequent recovery of benthic communities are dependent on their functional traits (Rumohr & Kujawski 2000, Blanchard et al. 2004, de Juan et al. 2007). The relationship between organism and sediment is also highly related to the vulnerability of species to disturbance.
Mortality resulting from fishing disturbance is usually size dependent, and it is widely accepted that there is a relationship between body size and life history (Jennings et al. 2001 a, de Juan et al. 2007). Large and slow-growing animals are correlated with lower annual reproductive output, making them more vulnerable to disturbance (Rumohr & Kujawski 2000). Thus, larger organisms like bivalves and echinoderms (which are among the highest contributors to total biomass) suffer very high mortality whereas smaller bivalves and polychaetes are less vulnerable because they are pushed aside with fluidized sediment generated by gear pressure (Bergman & Hup 1992, Gilkinson et al. 1998). Kaiser (1998) also reported that species presenting the greatest declines tend to be slow growing, to have low fecundity, or to be physically vulnerable to damage. In the current study, both smaller (<1 cm) and larger (1-5 cm) organisms were similarly affected by fishing. Mobility and position in sediment, together with the existence of external protecting structures and body design, are other major biological traits determining vulnerability of species to fishing. According to several studies, sedentary surface organisms are usually strongly affected (Bergman & Hup 1992, Thrush et al. 1995), whereas mobile or sedentary subsurface species that are able to avoid the fishing gear may be less affected (de Juan et al. 2007). In the current study, the most affected species presented low or no mobility, mainly with bodies of scales or chitinous plates, or even a vermiform shape. In terms of the existence of external protective structures, both tubiculous and nontubiculous species were similarly affected by fishing. Eleftheriou and Robertson (1992) reported a 50% reduction in tube-building polychaete species in an experiment undertaken in Scotland (10-m depth). In the current study, tubiculous species mainly correspond to amphipod and some polychaete species that dominated benthic samples before dredging. Their dominance confers to them the greatest potential for presenting variations after a general displacement of all species by the fishing gear, which may have conditioned the results.
[FIGURE 6 OMITTED]
Recovery of Macrobenthic Communities
In accordance with the general patterns found for most of variables analyzed in the current study, multivariate analysis supported the temporary effects of fishing between immediately after fishing and 3 days after fishing (depending on the variable used). However, for bivalves and gastropods, patterns were not as evident. These discrepancies may result from different sensitivities of the major taxonomic groups to the fishing impact that can be either maximized or attenuated when the entire community is analyzed together.
Considering that during the current experiment 4 days of consecutive fishing were undertaken in the HI area (i.e., 3 days after the first fishing event), and the recovery potential of benthic communities, it seems that the communities studied have high resilience to fishing. It is also worth noting that the main effects were not detected immediately after fishing. This may be related to the collection of injured organisms that might die during the ensuing hours. The high potential recovery of these communities could be related to: (1) the natural hydrodynamics of these areas and therefore the species present had been already subjected to a "natural selection," (2) a relatively quick return to the disturbed area of taxa removed by the fishing gear, or (3) colonization by specimens from nearby areas that were not impacted. At this depth (9 m), storm periods with high wave height are relatively frequent all year along the southern coast of Portugal (data not presented; gathered by Portuguese Hydrographic Institute). During the study period, particularly 20 days before the last sampling occasion, a major storm event lasted for 3 days with a maximum wave height of 4 m. This event was reflected in a general decrease of most variables in all areas and was not in multivariate analysis. The existence of frequent natural disturbance events in the study area may have shaped macrobenthic communities, with the dominant species being adapted to sediment disturbance. The absence of major fishing impacts in hydrodynamic areas had already been reported for the southern Portuguese coast (Constantino et al. 2009). Because fishing did not occur for long time in our study area, natural disturbance seems to be the most plausible reason for the results obtained, as well as the high recovery potential of the communities studied. Ragnarsson and Lindegarth (2009) also reported the dynamic nature of the environment where their study was conducted as a reason for the short-term effects of fishing observed. It is known that the degree of natural disturbance experienced by a community may determine the degree to which it is affected by fishing disturbance (Kaiser et al. 1996). Therefore, it is expected that, in the long term, the communities that live with high levels of natural disturbance are more resilient to anthropogenic disturbance, whereas communities from areas with low levels of natural disturbance are more susceptible (Queiros et al. 2006).
[FIGURE 7 OMITTED]
The assessment of impacts on marine communities caused by natural and human-induced changes is essential for management purposes. Nevertheless, the incorporation of extreme events like storms into experimental studies is not an option because they are unpredictable (Underwood & Chapman 2000). The major storm event that occurred in the current study (between 30 days and 120 days after fishing--specifically, 20 days before the last sampling occasion) showed that these natural disturbance events can have similar impacts to those of fishing. In particular, when the entire community is considered along with the two main taxonomic groups (Amphipoda and Polychaeta), the decrease in abundance and number of taxa was very pronounced. These samples appeared clearly separate from the remaining in the NMDS plots, resulting from a clear impoverishment in the benthic community. Unfortunately, the recovery of macrobenthic communities after the storm was not possible to assess, but because the sampling was undertaken several days after, and macrobenthic communities still showed signs of disturbance, we hypothesize that recovery is slower than that observed for fishing. The most probable reason is the extension of the impacted area, decreasing the effects of redistribution of benthic species through adjacent areas (Alves et al. 2003). Recolonization patterns of disturbed areas were already found to be size dependent (Smith & Brumsickle 1989, Thrush et al. 1996, Whitlatch et al. 1998). Smith and Brumsickle (1989) observed that postlarval stages were the main colonists for both smaller (50 [cm.sup.2]) and larger (1,750 [m.sup.2]) experimental plots, but their contribution was inversely related to the patch size, leading to faster colonization rates in smaller plots. Two main processes are involved in benthic recolonization: (1) larval settlement by species with planktonic dispersal mode (Butman 1987) and (2) passive or active movement of juvenile and adult specimens (Smith & Brumsickle 1989). Although the success of the first process depends primarily on the availability of larvae and the compatibility with sediment characteristics (Butman 1987), the latter is dependent on the availability of colonists in the surrounding areas and their capacity to move (Negrello Filho et al. 2006). Common to both processes are the interspecific interactions in the newly colonized habitat (Zajac et al. 1998). The relative importance of each process to the dynamics of marine assemblages depends on the characteristics of different habitats, and on the life histories and ecology of relevant species (Negrello Filho et al. 2006).
[FIGURE 8 OMITTED]
Despite the relative constancy of fishing effects on several measures of diversity, caution is needed when interpreting the results, because of the pseudo-replicated design applied. Indeed, several fishing impact studies suffers from pseudo-replication sensu Hurlbert (1984) or "confounding" sensu (Underwood 1997) as a result of the lack of replication, control areas, or before impact data (Pitcher et al. 2009). The nonexistence of several unfished areas along the southern Portuguese coast kept us from having a solid sampling design because of unreplicated treatments. Several additional aspects forced us to the current sampling design: (1) the relatively small area available for conducting the experiments, (2) the need to select areas in the same bathymetric line to avoid the influence of depth, (3) the need to guarantee that areas were separated enough to prevent interference between treatments, and (4) our intention to simulate closely real fishing operations. This is a classic case of pseudo-replication, which we tried to minimize (although it is impossible to eliminate) by assessing the small-scale variability in each treatment, by comparing treatments before fishing, and by analyzing the temporal patterns in the control area. Fishing effects were only analyzed when no one of these components were found to be not significant. These cautions when interpreting the results are of utmost importance because populations of different locations may change differently with time, regardless of any impact (Pitcher et al. 2009).
Lindegarth et al. (2000b) empirically addressed the question of adequate replication of treatments by comparing the results of a replicated BACI experiment (3 control sites vs. 3 impacted sites) with a series of unreplicated BACI experiments (1 control vs. 1 impacted). They found that spatial confounding may occur in unreplicated experiments. Based on the replicated experiment, however, any patterns of temporal change in measured variables could clearly be attributed to continuous trawling. Analyses of pairs of sites (1 control vs. 1 impacted) pointed out a large number of significant differences. Being aware of the constraints imposed by the current sampling design, there are some reasons that attenuate the design flaw and make us confident about the fishing impacts observed: (1) the minor significant small-scale variability found; (2) the relative temporal constancy of the control area, with the main differences detected between poststorm (120 days) and prestorm periods; (3) the almost nonexistent differences before fishing between treatment areas; and (4) the resemblance of the current findings with other studies using sampling design.
The integration of the information gathered during this study pointed to a significant short-term decrease (up to 3 days after fishing) of most community variables, with clear signs of macrobenthic recovery being observed thereafter. The use of different community variables and methods when analyzing the data set was found to be useful, providing complementary information and allowing for a more comprehensive knowledge of the hypotheses being tested. The occurrence of a major storm event 20 days before the last sampling occasion, and the signs of disturbance detected on macrobenthic communities from all selected areas during this last sampling period, allowed us to infer that potential effects resulting from these natural events can cause more serious impacts than those resulting from the experimental fishing operations. The relatively frequent natural disturbing events in the studied area may have shaped macrobenthic communities, conferring resilience to them for stressful episodes like those resulting from fishing.
We thank the crews of RV Diplodus for their skilful help. We are also grateful to Dr. M. N. Santos, J. Curdia, and T. Simoes for their help during fieldwork, as well as to Clube Naval de Portimao for logistical support. We are indebted to Dr. O. Ferreira (CIACOMAR) for sediment sample processing and sediment characterization, and to Dr. P. Range (CCMAR) for his assistance in statistical analysis. Sincere thanks are also due to one anonymous referee whose suggestions greatly improved the manuscript. This investigation was funded by the TIMES project (Towards Integrated Management of Ensis Stocks) under the INTERREG IIIB Programme--Atlantic Area and European Union--FEDER Programme. S.C. (SFRH/BPD/26986/2006) benefited from a postdoctoral grant from FCT (Fundacao para a Ciencia e a Tecnologia).
Alves, F., L. Chicharo, A. Nogueira & J. Regala. 2003. Changes in benthic community structure due to clam dredging on the Algarve coast and the importance of seasonal analysis. J. Mar. Biol. Assoc. UK 83:719-729.
Bergman, M. J. N., B. Ball, C. Bijleveld, J. A. Craeymeersch, B. W. Munday, H. Rumohr & J. W. Van Santbrink. 1998. Direct mortality due to trawling. In: H. J. Lindeboom & S. J. de Groot, editors. The effects of different types of fisheries on the North Sea and Irish Sea benthic ecosystems. Nioz-Rapport 1998-1/Rivo-Dlo report C003/ 98. pp. 167-185.
Bergman, M. J. N. & M. Hup. 1992. Direct effects of beam trawling on macrofauna in a sandy sediment in the southern North Sea. ICES J. Mar. Sci. 49:5-11.
Blanchard, F., F. LeLoc'h, C. Hily & J. Boucher. 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. 280:249-260.
Blott, S. J. & K. Pye. 2001. GRADISTAT: a grain size distribution and statistics package for the analysis of unconsolidated sediments. Earth Surf Process. Landf. 26:1237-1248.
Bradshaw, C., L. O. Veale & A. R. Brand. 2002. The role of scallop-dredge disturbance in long-term changes in Irish Sea benthic communities: a re-analysis of an historical dataset. J. Sea Res. 47:161-184.
Bremner, J., S. I. Rogers & C. L. J. Frid. 2003. Assessing functional diversity in marine benthic ecosystems: a comparison of approaches. Mar. Ecol. Prog. Ser. 254:11-25.
Bustos-Baez, S. & C. Frid. 2003. Using indicator species to assess the state of macrobenthic communities. Hydrobiologia 496:299-309.
Butman, C. A. 1987. Larval settlement of soft-sediment invertebrates: the spatial scales of pattern explained by active habitat selection and the emerging role of hydrodynamical processes. Oceanogr. Mar. Biol. Annu. Rev. 25:113-165.
Cardoso, P., M. Bankovic, D. Raffaelli & M. A. Pardal. 2007. Polychaete assemblages as indicators of habitat recovery in a temperate estuary under eutrophication. Estuar. Coast. Shelf Sci. 71:301-308.
Chicharo, L., M. A. Chicharo, M. Gaspar, F. Alves & J. Regala. 2002a. Ecological characterization of dredged and recently dredge-free areas of a bivalve fishing ground, off South Portugal. J. Mar. Biol. Assoc. UK 82:41-50.
Chicharo, L., A. Chicharo, M. B. Gaspar, J. Regala & F. Alves. 2002b. Reburial time and indirect mortality of Spisula solida clams caused by dredging. Fish. Res. 59:247-257.
Clarke, K. R. & R. M. Warwick. 1994. Change in marine communities: an approach to statistical analysis and interpretation. Plymouth: Natural Environmental Research Council. 144 pp.
Collie, J. S., S. J. Hall, M. J. Kaiser & I. R. Poiners. 2000. A quantitative analysis of fishing impacts on shelf sea benthos. J. Anim. Ecol. 69:785-798.
Constantino, R., M. B. Gaspar, J. Tata-Regala, S. Carvalho, J. Curdia, T. Drago, R. Taborda & C. C. Monteiro. 2009. Clam dredging effects and subsequent recovery of benthic communities at different depth ranges. Mar. Environ. Res. 67:89-99.
Dayton, P. K., S. F. Thrush, M. T. Agardy & R. F. Hofman. 1995. Environmental effects of marine fishing. Aquat. Conserv. 5:205-232.
de Juan, S., S. F. Thrush & M. Demestre. 2007. Functional changes as indicators of trawling disturbance on a benthic community located in a fishing ground (NW Mediterranean Sea). Mar. Ecol. Prog. Ser. 334:117-129.
Eleftheriou, A. & M. R. Robertson. 1992. The effects of experimental scallop dredging on the fauna and physical environment of a shallow sandy community. Neth. J. Sea Res. 30:289-299.
Ellingsen, K. E., K. R. Clarke, P. J. Somerfield & R. M. Warwick. 2005. Taxonomic distinctness as a measure of diversity applied over a large scale: the benthos of the Norwegian continental shelf. J. Anita. Ecol. 74:1069-1079.
Folk, R. L. 1954. The distinction between grain size and mineral composition in sedimentary rock nomenclature. J. Geol. 62:344-359.
Folk, R. L. & W. C. Ward. 1957. Brazos River bar: a study in the significance of grain size parameters. J. Sed. Petrol. 27:3-26.
Frid, C. L. J., R. A. Clark & J. A. Hall. 1999. Long-term changes in the benthos on a heavily fished ground off the NE coast of England. Mar. Ecol. Prog. Ser. 188:13-20.
Frid, C. L. J., K. G. Harwood, S. J. Hall & J. A. Hall. 2000. Long-term changes in the benthic communities on the North Sea fishing grounds. ICES J. Mar. Sci. 57:1303-1309.
Garcia, E. G., S. A. Ragnarsson & H. Eiriksson. 2006. Effects of scallop dredging on macrobenthic communities in West Iceland. ICES J. Mar. Sci. 63:434-443.
Gaspar, M. B., S. Carvalho, R. Constantino, J. Tata-Regala, J. Curdia & C. C. Monteiro. 2009. Can we infer dredge fishing effort based on macrobenthic community structure? ICES J. Mar. Sci. 66:2121-2132.
Gaspar, M. B., M. Castro & C. C. Monteiro. 1999. Effect of tooth spacing and mesh size on the catch of the Portuguese clam and razor clam dredge. ICES J. Mar. Sci. 56:103-110.
Gaspar, M. B. & L. M. Chicharo. 2007. Modifying dredges to reduce by-catch and impacts on the benthos. In: S. J. Kennelly, editor. By-catch reduction in the world's fisheries, vol. 7. Netherlands: Springer. pp. 95-140.
Gaspar, M. B., F. Leitao, M. N. Santos, L. Chicharo, M. D. Dias, A. Chicharo & C. C. Monteiro. 2003a. A comparison of direct macro-faunal mortality using three types of clam dredges. ICES J. Mar. Sci. 60:733-742.
Gaspar, M. B., M. N. Santos, F. Leitao, L. Chicharo, A. Chicharo & C. C. Monteiro. 2003b. Recovery of substrates and macro-benthos after fishing trials with a new Portuguese clam dredge. J. Mar. Biol. Assoc. UK 83:713-717.
Gaston, G. R., C. F. Rakocinski, S. S. Brown & C. M. Cleveland. 1998. Trophic function in estuaries: response of macrobenthos to natural and contaminant gradients. Mar. Freshw. Res. 49:833-846.
Gaudencio, M. J. & H. N. Cabral. 2007. Trophic structure of macrobenthos in the Tagus estuary and adjacent costal shelf. Hydrobiologia 587:241-251.
Gilkinson, K. D., G. B. J. Fader, D. C. Norton, Jr., R. Charron, D. Mckeown, D. Roddick, E. L. R. Kenchington, K. MacIsaac, C. Bourbonnais, P. Vass & Q. Liu. 2003. Immediate and longer-term impacts of hydraulic clam dredging on an offshore sandy seabed: effects on physical habitat and processes of recovery. Cont. Shelf Res. 23:1315-1336.
Gilkinson, K. D., D. C. Gordon, Jr., K. G. MacIsaac, D. L. Mckeown, E. L. R. Kenchinton, C. Bourbonnais & W. P. Vass. 2005. Immediate impacts and recovery trajectories of macrofaunal communities following hydraulic clam dredging on Banquereau, eastern Canada. ICES J. Mar. Sci. 62:925-947.
Gilkinson, K., M. Paulin, S. Hurley & P. Schwinhamer. 1998. Impact of trawl door scouring on infaunal bivalves: results of a physical trawl door/model/dense interaction. J. Exp. Mar. Biol. Ecol. 224:291-312.
Hall, S. J. 1994. Physical disturbance and marine benthic communities: life in unconsolidated sediments. Oceanogr. Mar. Biol. Annu. Rev. 32:179-239.
Hall, S. J., D. J. Basford & M. R. Robertson. 1990. The impact of hydraulic dredging for razor clams Ensis sp. on an infaunal community. Neth. J. Sea Res. 27:119-125.
Hauton, C., R. J. A. Atkinson & P. G. Moore. 2003. The impact of hydraulic blade dredging on a benthic megafaunal community in the Clyde Sea area, Scotland. J. Sea Res. 50:45-56.
Hurlbert, S. J. 1984. Pseudoreplication and the design of ecological field experiments. Ecol. Monogr. 54:187-211.
Jenkins, S. R., B. D. Beukers-Stewart & A. R. Brand. 2001. Impact of scallop dredging on benthic megafauna: a comparison of damage levels in captured and non-captured organisms. Mar. Ecol. Prog. Ser. 215:297-301.
Jennings, S., T. A. Dinmore, D. E. Duplisea, K. J. Warr & J. E. Lancaster. 2001a. Trawling disturbance can modify benthic production processes. J. Anim. Ecol. 70:459-475.
Jennings, S. & M. J. Kaiser. 1998. The effects of fishing on marine ecosystems. Adv. Mar. Biol. 34:201-252.
Jennings, S., J. K. Pinnegar, V. C. Polunin & K. J. Warr. 2001b. Impacts of trawling disturbance on the trophic structure of benthic invertebrate communities. Mar. Ecol. Prog. Ser. 213:127-142.
Kaiser, M. J. 1998. Significance of bottom-fishing disturbance. Conserv. Biol. 12:1230-1235.
Kaiser, M. J., K. R. Clarke, H. Hinz, M. C. V. Austen, P. J. Somerfield & I. Karakassis. 2006. Global analysis of response and recovery of benthic biota to fishing. Mar. Ecol. Prog. Ser. 311:1-14.
Kaiser, M. J., A. S. Hill, K. Ramsay, B. E. Spencer, A. R. Brand, L. O. Veale, K. Prudden, E. I. S. Rees, B. W. Munday, B. Ball & S. J. Hawkins. 1996. Benthic disturbance by fishing gear in the Irish Sea: a comparison of beam trawling and scallop dredging. Aquat. Conserv. 6:269-285.
Kaiser, M. J., H. Hinz, R. M. Callaway, A. Nall & C. L. Biles. 2005. Resource degradation: a subtle effect of bottom fishing. Mar. Biol. 146:401-408.
Kaiser, M. J., K. Ramsay, C. A. Richardson, F. E. Spence & A. R. Brand. 2000. Chronic fishing disturbance has changed shelf sea benthic community structure. J. Anim. Ecol. 69:494-503.
Kaiser, M. J. & B. E. Spencer. 1994. Fish scavenging behaviour in recently trawled areas. Mar. Ecol. Prog. Ser. 112:41-49. Kaiser, M. J. & B. E. Spencer. 1996. The effects of beam-trawl disturbance on infaunal communities in different habitats. J. Anim. Ecol. 65:348-358.
Koslow, J. A., K. Gowlett-Holmes, J. K. Lowry, T. O'Hara, G. C. B. Poore & A. Williams. 2001. Seamount benthic macrofauna off southern Tasmania: community structure and impacts of trawling. Mar. Ecol. Prog. Ser. 213:111-125.
Lenihan, H. S., C. H. Peterson, S. L. Kim, K. E. Conlan, R. Fairey, C. McDonald, J. H. Grabowski & J. S. Oliver. 2003. Variation in marine benthic community composition allows discrimination of multiple stressors. Mar. Ecol. Prog. Ser. 261:63-73.
Levin, L. A. 2000. Polychaetes as environmental indicators: response to low oxygen and organic enrichment. Bull. Mar. Sci. 67:668-678.
Lindeboom, H. J. & S. J. de Groot. 1998. The effects of different types of fisheries on the North Sea and Irish Sea benthic ecosystems: NIOZ rapport 1998-1. Texel: Netherlands Institute for Sea Research (NIOZ). 404 pp.
Lindegarth, M., D. Valentinsson, M. Hansson & M. Ulmestrand. 2000a. Effects of trawling disturbances on temporal and spatial structure of benthic soft-sediment assemblages in Gullmarsfjorden, Sweden. ICES J. Mar. Sci. 57:1369-1376.
Lindegarth, M., D. Valentinsson, M. Hansson & M. Ulmestrand. 2000b. Interpreting large-scale experimental effects of trawling on benthic fauna: an empirical test of the potential effects of spatial confounding in experiments without replicated control and trawled areas. J. Exp. Mar. Biol. Ecol. 245:155-169.
Mancinelli, G., S. Fazi & L. Rossi. 1998. Sediment structural properties mediating dominant feeding type patterns in soft-bottom macrobenthos of the northern Adriatic Sea. Hydrobiologia 367:211-222.
Morello, E. B., C. Froglia, R. J. A. Atkinson & P. G. Moore. 2006. Medium-term impacts of hydraulic clam dredgers on a macrobenthic community of the Adriatic Sea (Italy). Mar. Biol. 149:401-413.
Negrello Filho, O. A., A. J. Underwood & M. G. Chapman. 2006. Recolonization of infauna on a tidal flat: an experimental analysis of modes of dispersal. J. Exp. Mar. Biol. Ecol. 328:240-250.
Pitcher, C. R., C. Y. Burridge, T. J. Wassenberg, B. J. Hill & I. R. Poiner. 2009. A large scale BACI experiment to test the effects of prawn trawling on seabed biota in a closed area of the Great Barrier Reef Marine Park, Australia. Fish. Res. 99:168-183.
Pranovi, F., F. Da Ponte, S. Raicevich & O. Giovanardi. 2004. A multidisciplinary study of the immediate effects of mechanical clam harvesting in the Venice Lagoon. ICES J. Mar. Sci. 61:43-52.
Queiros, A. M., J. G. Hiddink, M. J. Kaiser & H. Hinz. 2006. Effects of chronic bottom trawling disturbance on benthic biomass production and size spectra in different habitat's disturbance. J. Exp. Mar. Biol. Ecol. 335:91-103.
Ragnarsson, S. A. & M. Lindegarth. 2009. Testing hypotheses about temporary and persistent effects of otter trawling on infauna: changes in diversity rather than abundance. Mar. Ecol. Prog. Ser. 385:51-64.
Ramsay, K., M. J. Kaiser, P. G. Moore & R. N. Hughes. 1997. Consumption of fisheries discards by benthic scavengers: utilization of energy subsidies in different marine habitats. J. Anita. Ecol. 66:884-896.
Roth, S. & J. G. Wilson. 1998. Functional analysis by trophic guilds of macrobenthic community structure in Dublin Bay, Ireland. J. Exp. Mar. Biol. Ecol. 222:195-217.
Rumohr, H. & T. Kujawski. 2000. The impact of trawl fishery on the epifauna of the southern North Sea. ICES J. Mar. Sci. 57:1389-1394.
Schratzberger, M., T. A. Dinmore & S. Jennings. 2002. Impacts of trawling on the diversity, biomass and structure of meiofauna assemblages. Mar. Biol. 140:83-93.
Smith, C. R. & S. J. Brumsickle. 1989. The effects of patch size and substrate isolation modes and rates in an intertidal sediment. Limnol. Oceanogr. 34:1263-1277.
Sprung. M. 1994. Macrobenthic secondary production in the intertidal zone of the Ria Formosa: a lagoon in southern Portugal. Estuar. Coast. Shelf Sci. 38:539-558.
Suguio, K. 1973. Introducao a sedimentologia. Sao Paulo: Edgard Blucher. 317 pp.
Thrush, S. F., J. E. Hewitt, V. J. Cummings & P. K. Dayton. 1995. The impact of habitat disturbance by scallop dredging on marine benthic communities: what can be predicted from the results of experiments? Mar. Ecol. Prog. Ser. 129:141-150.
Thrush, S. F., R. B. Whitlatch, R. D. Pridmore. J. E. Hewitt, V. J. Cummings & M. R. Wilkinson. 1996. Scale-dependent recolonization: the role of sediment stability in a dynamic sandflat habitat. Ecology 77:2472-2487.
Tuck, I. D., N. Bailey, M. Harding, G. Sangster, T. Howell, N. Graham & M. Breen. 2000. The impact of water jet dredging for razor clams, Ensis spp., in a shallow sandy subtidal environment. J. Sea Res. 43:65-81.
Tuck, I. D., S. J. Hall, M. R. Robertson, E. Armstrong & D. J. Basford. 1998. Effects of physical trawling disturbance in a previously unfished sheltered Scottish sea loch. Mar. Ecol. Prog. Ser. 162:227-242.
Turner, S. J., S. F. Thrush, J. E. Hewitt, V. J. Cummings & G. Funnell. 1999. Fishing impacts and the degradation or loss of habitat structure. Fish. Manage. Ecol. 6:401-420.
Underwood, A. J. 1997. Experiments in ecology: their logical design and interpretation using analysis of variance. Cambridge: Cambridge University Press. 504 pp.
Underwood, A. J. & M. G. Chapman. 2000. Variation in abundances of intertidal populations: consequences of extremities of environment. Hydrobiologia 426:25-36.
Veale, L. O., A. S. Hill, S. J. Hawkins & A. R. Brand. 2000. Effects of long-term physical disturbance by commercial scallop fishing on subtidal epifaunal assemblages and habitats. Mar. Biol. 137:325-337.
Warwick, R. M. & K. R. Clarke. 1995. New "biodiversity" measures reveal a decrease in taxonomic distinctness with increasing stress. Mar. Ecol. Prog. Ser. 129:301-305.
Wentworth, C. K. 1922. A scale of grade and class terms for clastic sediments. J. Geol. 30:377-392.
Whitlatch, R. B., A. M. Lohrer, S. F. Thrush, R. D. Pridmore, J. E. Hewitt, V. J. Cummings & R. N. Zajac. 1998. Scale-dependent benthic recolonization dynamics: life stage-based dispersal and demographic consequences, Hydrobiologia 375/376:217-226.
Zajac, R. N., R. B. Whitlatch & S. F. Thrush. 1998. Recolonization and succession in soft-sediment infaunal communities: the spatial scale of controlling factors. Hydrobiologia 375/376:227-240.
SUSANA CARVALHO, (1) RITA CONSTANTINO, (1) FABIO PEREIRA, (1) RADHOUAN BEN-HAMADOU (2) AND MIGUEL B. GASPAR (1) *
(1) National Institute of Biological Resources (INRB, I.P.)/IPIMAR, Av. 5 de Outubro s/n, 8700-305 Olhao, Portugal; (2) Centre of Marine Sciences (CIMAR/CCMAR), EcoReach Research Group, University of Algarve, Campus de Gambelas, 8005-139 Faro, Portugal
* Corresponding author. E-mail: email@example.com
TABLE 1. Results of nested ANOVA for abundance (N) and number of taxa (S) for all taxa and main taxonomic groups separately, Margalef species richness (d), Shannon-Wiener (H), Pielou's equitability (J) and biomass, average taxonomic distinctness ([[DELTA].sup.+]) for all community and main taxonomic groups polychaete families, and the ratio between carnivores and suspension feeders (C/SF). Site Time X Variable Time Treatment (Treatment) Treatment N All taxa *** ns ns *** Amphipoda *** ** ns *** Polychaeta *** ns ns ** Gastropoda *** *** P > 0.25 ns Bivalvia *** * P > 0.25 ** S All taxa *** ** ns *** Amphipoda *** * * *** Polychaeta *** ** ns * Gastropoda *** ns P > 0.25 ns Bivalvia ** ns P > 0.25 * d *** * ns *** H' ** * ns *** J' ns *** * ** Biomass * * ns * [[DELTA].sup.+] All taxa *** ns ns ns Crustacea *** ns P > 0.25 ** Annelida ns ns ns ns Mollusca * ns ns ns Spionidae *** *** P > 0.25 *** Opheliidae *** ** P > 0.25 *** Capitellidae ns ns ns ns Terebellidae *** *** P > 0.25 * Sabellidae ns * P > 0.25 ns Pectinariidae ns * P > 0.25 ns Goniadidae * ns ** ns Hesionidae *** ns ns * C/SF *** ne P > 0 Time X Site Variable (Treatment) Cochran's N All taxa ns ** Amphipoda ns ns Polychaeta ns ** Gastropoda ns ns Bivalvia ns ns S All taxa ns ns Amphipoda * ns Polychaeta ns ns Gastropoda P > 0.25 ns Bivalvia ns ns d ns ns H' ns ns J' *** ** Biomass ns ns [[DELTA].sup.+] All taxa ns ns Crustacea ns ns Annelida ns ** Mollusca ns ns Spionidae ns ns Opheliidae P > 0.25 ns Capitellidae ns ** Terebellidae ** ns Sabellidae ns ns Pectinariidae ns ns Goniadidae * ns Hesionidae *** ns C/SF Time and treatment are orthogonal fixed factors. Site is a random factor nested in treatment. * P < 0.05, ** P < 0.01, *** P < 0.001; ns, 0.05 < P < 0.25. TABLE 2. Post hoc comparisons for the interactive effect time X treatment. FT B IA 1 day 2 days Protection N ns ns C > MI *** C > HI * T ns ns C > MI = HI *** C > MI = HI ** Size (cm) <1 ns ns C > MI = HI *** C > MI = HI ** 1-5 ns ns C > MI = HI * C > HI ** Body design S/CP ns C > HI * C > MI = HI *** C > MI = HI ** S ns ns C > MI ** ns V ns ns C > MI = HI *** C > MI = HI ** Mobility Sed ns ns C > MI = HI *** C > HI *** Low ns ns C > MI = HI *** C > MI * > HI * Med ns ns ns C > MI = HI ** FT 3 days 7 days 14 days 30 days 120 days Protection N C > MI = HI ** ns ns ns ns T C > HI ** ns ns ns ns Size (cm) <1 C > MI = HI ** ns ns ns ns 1-5 C > MI = HI *** ns ns ns ns Body design S/CP C > MI = HI *** ns ns ns ns S ns ns ns ns ns V C = MI > HI ** ns ns ns ns Mobility Sed C > HI * ns ns ns ns Low C > MI ** ns ns ns ns Med ns ns ns ns ns B, before; C, control; FT, functional trait; HI, highly impact; IA, immediately after fishing; low, low mobility; Med, medium mobility; MI, moderately impacted; N, no external protection; ns, not significant; S, shell; S/CP, scales or chitinous plates; Sed, sedentary; T, with external protection; V, vermiform. * P < 0.05, ** P < 0.01, *** P < 0.001.
|Printer friendly Cite/link Email Feedback|
|Author:||Carvalho, Susana; Constantino, Rita; Pereira, Fabio; Ben-Hamadou, Radhouan; Gaspar, Miguel B.|
|Publication:||Journal of Shellfish Research|
|Date:||Aug 1, 2011|
|Previous Article:||Validation of age estimation in geoduck clams using the bomb radiocarbon signal.|
|Next Article:||Microsatellites of Mytilus chilensis: a genomic print of its taxonomic status within Mytilus sp.|