Previous Page Table of Contents Next Page


Biological Impacts


An overview of the impact studies included in this review is given in Tables 2 to 4. The following section on otter trawling on soft bottom habitats is organized as follows: high seas (offshore) fishing grounds, shrimp and Nephrops trawling, and studies in the Mediterranean. These three groupings represent different types of fisheries and habitats in which similar and comparable studies have been conducted.

OTTER TRAWLING ON SOFT BOTTOM HABITATS

Experimental trawling on sandy bottoms of high seas fishing grounds caused short-term declines in some individual taxa, but did not produce large changes in the benthic assemblage. Natural variability was found to be considerable, and these habitats may be resistant to trawling owing to high degrees of natural disturbance.

One of the most comprehensive experiments on the impacts of otter trawling is the three-year study conducted on the Grand Banks of Newfoundland (Prena et al., 1999; Kenchington et al., 2001). The northeastern part of this fishing ground has a rich, diverse and homogeneous benthic community and a bottom habitat that is representative of large areas off the coast of Atlantic Canada. Fishing effort records indicate that this location had not been fished since the early 1980s, and the region was closed to all fishing activity in 1992 as a result of the collapse of the groundfish stocks (Kenchington et al., 2001), so this site fulfils several important requirements of an impact study. Three experimental corridors with parallel reference corridors were established. The experimental corridors were trawled 12 times within a five-day period for three successive years starting in 1993. This trawling disturbance must be considered to be of high intensity. The effects of the experimental trawling on the mega-epibenthic community, as sampled with epibenthic sledge, are reported by Prena et al. (1999), and the effects on the macrofaunal community, as sampled by a large video-equipped grab, are reported by Kenchington et al. (2001). A before/after control/impact (BACI) design was attempted, but the intended sledge sampling programme was considerably reduced, thus the resulting mega-epibenthic samples did not allow examination of the recovery of these taxa.

The sledge samples of mega-epibenthic species demonstrated that the trawling disturbance caused an average decrease in total biomass of 24 percent in trawled corridors. This decrease was owing primarily to reductions in biomass of sand dollars, brittle stars, soft corals, sea urchins and snow crabs, whereas no significant effects were observed for mollusc species. Scavenger predation on dead or damaged organisms and change in catchability of the sampling sledge (owing to the burial of organisms by resuspended sediment) were suggested as the most likely factors causing lower biomass in the trawled corridors. Trawling caused significant physical damage only on the echinoderms (sand dollar, brittle star, sea urchin), with the greatest probability of impact on the sea urchin (10 percent damage). The study indicated that trawling caused decreased homogeneity and increased aggregation of mega-epibenthic organisms immediately after trawling. Although the sampling programme was not designed to determine long-term effects and recovery, available evidence indicated that the habitat and biological community recovered from the annual trawling disturbance in a year or less. Total biomass showed considerable interannual variability.

Considerable temporal variability in the community properties of reference corridors was also observed for the grab samples of the macrofauna (Kenchington et al., 2001). These samples demonstrated that the total number of species and the total abundance in both reference and trawled corridors declined during the three-year experiment. The large number of species that decreased in abundance (43 taxa) and biomass (35 taxa) showed that the macrofaunal community in the reference corridors experienced considerable change through the course of the study, indicating that the benthic community at the study site is naturally dynamic and exhibits temporal changes irrespective of trawling disturbance. The mean total abundance per grab sample was lower (25 percent) immediately after trawling in one of the three years of the experiment. This decline in abundance was demonstrated for 13 taxa (mostly polychaetes), of which 11 also declined in biomass. These taxa seemed to recover within a year or less. None of the other community indices in any of the years showed an immediate significant effect of trawling, and few community indices and individual taxa showed a significant long-term effect of trawling. The authors concluded that a natural long-term decline in the macrofaunal assemblage (total numbers of species and individuals) was the most prominent feature of the study.

The impacts of experimental trawling have also been studied on a high seas (offshore) fishing ground in the Barents Sea that consisted of silt/sand/gravel mixed with shell fragments (Kutti et al., in press). The research site was located within the Fishery Protection Zone around Bear Island, which was closed to commercial fishing activity in 1978 in order to protect juvenile gadoids. Five parallel transects were established, of which one was intensively trawled (700 percent coverage by the area of the door spread), one was moderately trawled (230 percent coverage) and three were used as controls. Sampling by video-equipped sledge was conducted before and immediately after trawling and every six months for 18 months after the trawling disturbance. To date, the only samples to be analysed and reported are those that were taken before and immediately after trawling in the intensively trawled transect and in one of the control transects (Kutti et al., in press.).

Trawling affected the benthic assemblage mainly through resuspension of surface sediment and displacement to the surface of shallow-burrowing infaunal bivalves. Thus, significant increases in the abundance and biomass of a majority of the infaunal bivalves and some burrowing gastropods were found after trawling. Although crustaceans as a group did not show a consistent response to trawling, the abundance of some species showed a significant reduction, probably owing to their ability to move out of the disturbed area. Diversity based on biomass data increased after trawling. Few individuals with physical damage were found in the sledge samples. It was concluded that the experimental trawling did not cause large changes in the benthic assemblage, which thus seemed to be resistant to the disturbance imposed. This resistance to trawling was explained by the high degree of natural disturbance (strong current, large temperature fluctuations) in the area.

In a study conducted in the eastern Bering Sea, previously unfished (UF) and heavily fished (HF) areas straddling a closed-area boundary were compared in order to investigate the long-term consequences for the benthos (McConnaughey, Mier and Dew, 2000). A total of 42 pairs of UF and HF sites were sampled by two trawlers using a modified otter trawl. A significant difference in biomass was found between HF and UF treatments, although the direction of this difference was not consistent for the different taxa or for the two separate sets of samples taken by the two vessels. Several taxa were considered, but a significant difference in biomass was found for only a few species. One data set showed that Actiniaria and Neptunea were significantly more abundant in the UF area, and the other data set showed that this area had lower abundances of empty gastropod shells, snail eggs and Porifera. Diversity was significantly lower in the HF area, as a result of greater dominance by the sea star (Asterias amurensis) in this area. Among sedentary organisms, niche breadth was also lower in the HF area, indicating a more patchy distribution for the attached or non-motile species in the HF area. The generalized conclusions drawn were reduced biomass (sponges and anemones), niche breadth and diversity of sedentary macrofauna in the HF area, and mixed responses within motile groups and infaunal bivalves.

In this study, the power of many of the statistical tests was extremely low; hence there was little chance of detecting a difference between the UF and HF areas. In order to decrease the probability of type II error and increase the probability of correctly rejecting the null hypothesis of no difference, the authors used a significance level of 0.10 (several p-values were in the 0.05 to 0.10 range). The sampling gear (a modified sampling trawl) used can be regarded, at best, as semi-quantitative for benthic fauna.

Owing to a lack of true or replicate control sites, the changes in benthic assemblages demonstrated in some impact studies may reflect natural variability (spatial or temporal) and not the effects of trawling disturbance.

Owing to the lack of true control areas on the west coast of California (United States), the study of Engel and Kvitek (1998) used the fishing pressure gradient between two fishing grounds as a basis for comparison. The highly trawled site was trawled an average of four times per year (1987 to 1992, trawl logbook records), compared with once every three years for the lightly trawled site. Densities of epifauna (> 5 cm) were estimated from video transects conducted in 1994, and three years of grab samples were collected for infaunal analysis (1994 to 1996). Densities of epifaunal species were higher in the lightly than in the highly trawled area, and the difference was significant for sea pens (Ptilosarcus sp.), sea stars (Mediaster sp.), sea anemones (Urticina sp.) and sea slugs (Pleurobranchaea californica). The analysis of infaunal species showed significantly more polychaete species in the lightly trawled area, but no difference for crustacean species. The densities of oligochaetes and nematodes were higher in the highly than in the lightly trawled area, but the difference was significant for only one of the three years of the study. The polychaete Chloeia pinnata was significantly more abundant and had higher biomass in the highly trawled area in two of the years. In summary, the study indicated that intensive trawling decreased habitat heterogeneity (fewer rocks and mounds), epifauna density, number of polychaete species and favoured opportunistic species (oligochaetes and nematodes). However, the authors stated that some of these differences may have been attributable to physical differences between sites because there were no unfished control sites, no site replication and the sample sizes were small.

The study site of the investigation by Tuck et al. (1998) used to be a good fishing ground, but it had been closed to fishing for more than 25 years prior to the experimental trawling owing to the presence of a naval base. The treatment area was trawled on one day each month for 16 months, and ten trawl hauls were made on each trawling day. This intensive trawling activity was conducted by a trawl with no net. Faunal sampling was conducted prior to and during the trawl disturbance period (after five, ten and 16 months of disturbance) and after six, 12 and 18 months of recovery. Infauna was sampled using a grab, and epifaunal species were observed with a camera mounted on a sledge.

The study demonstrated an increased number of infaunal species and individuals in the trawled area during the period of disturbance, but no effect on biomass. Measures of diversity and evenness decreased in the trawled area. Two species of polychaetes were found to increase while two other polychaete species, although less obvious, decreased in abundance in response to trawl disturbance. Most of the changes that occurred after disturbance were reversed after trawling ceased (number of individuals, diversity, evenness), but some differences (number of species) between the treatment and reference sites remained throughout the 18-month recovery period. These measures changed over time at both sites, suggesting seasonal and annual fluctuations in the community studied.

The results of this study should be interpreted with caution because it had several deficiencies. In order not to reduce the epibenthic scavenger populations, which are potentially important mortality agents for exposed benthic fauna, their densities were preserved by using a trawl with no net. The authors claimed that this modification reduced the weight of the gear, and therefore they added extra weight (mass not given) to the ground rope. No studies have been conducted to determine the effect of removing the net of a trawl, but flume tank observations indicated that such a modification causes increased ground contact because the net lessens the weight of the ground rope during a tow (A. Engås, personal communication). Samples were taken from a single treatment and reference area, so this design is spatially confused and not suitable for demonstrating treatment effects (see Chapter 2 Methodologies). However, the authors argue that the effects of disturbance were examined by sampling at repeated points during a period of impact and recovery. In addition to this deficiency, it should be noted that before the experimental trawling there were significant differences between the treatment and reference sites in sediment grain size, sea bed roughness, organic carbon level and proportional abundance of the dominant phyla (polychaetes and molluscs), and there were also differences in several of the measures used to demonstrate the impacts of trawl disturbance (e.g. total number of individuals, abundance of some species, diversity and evenness).

An interesting approach was applied to evaluating the disturbances caused by flounder trawling in an intertidal habitat (Brylinsky, Gibson and Gordon Jr., 1994). The macrotidal character of the Bay of Fundy (Nova Scotia, Canada) afforded a unique sampling technique in that experimental trawling was conducted at high water, while biological samples were easily obtained at low water when the trawl tracks were exposed. The dominant groups of benthic organisms at the study site were nematodes and polychaete worms. After trawling, nematodes were less abundant in the door furrows than in the control samples, but this difference levelled out after four to six weeks. Trawling had no impact on either species composition or number of polychaetes, which were predominantly tube-dwelling or burrowing. This study indicated that the impact of flounder trawling is minor on an intertidal benthic community that is exposed to natural processes such as wave action and storms and that thus represents a heavily stressed environment.

Several studies have addressed the impacts of shrimp trawling on clayey-silt bottoms. No clear and consistent effects attributable to trawling were detected. Potential disturbance effects may be masked by the more pronounced temporal variability demonstrated in these studies.

The Gullmarsfjord on the Swedish west coast was closed to fishing for six and a half years before an experiment designed to evaluate the potential consequences of reintroducing the shrimp-trawl fishery was carried out (Hansson et al., 2000; Lindegarth et al., 2000a). Three pairs of treatment and control sites were arranged along the fjord. The treatment sites were trawled once a week (two hauls) over one year, giving a conservative estimate of 24 passages of the trawl over any given area within the trawled sites. The treatment sites were thus intensively trawled, but the gear used was a small and light shrimp trawl. Each site was sampled with a grab four times over a four-month period before trawling and another four times after trawling started. To date, this is the only experiment assessing the impacts of long-term trawling to be based on a comprehensive sampling programme with replicate treatment and control sites.

Biomass and the total number of individuals were shown to decrease during the experiment, but this effect could not be attributed to trawl disturbance, although the magnitude of the decreases were higher in the trawled than in the control sites (Hansson et al., 2000). However, the number of echinoderms decreased consistently in the trawled sites, whereas there was no change in the control sites. This significant decrease in the number of echinoderms (30 percent) was mainly reflected among the brittle stars (in particular the species Amphiura chiajei). Abundances of other phyla were also shown to change, but because changes were demonstrated to occur before trawling in the control sites and among sites within treatments, these differences could not be interpreted as the effects of trawling.

The authors discuss contradictory results between their study and previous experiments. The negative effect of trawling on echinoderm abundance was mainly owing to a decrease in the abundance of ophiuroids, which have been shown to be resistant to, or even favoured by, trawling in other studies (Lindley, Gamble and Hunt, 1995; Kaiser and Spencer, 1996; Tuck et al., 1998). The hypothesis that the ratio between molluscs and polychaetes will decrease as a consequence of trawling (Thrush et al., 1998) was not supported by this study. Furthermore, the results of this study differed from those of most other manipulative experiments in that fewer taxa appeared to be affected by trawling. The authors give several possible explanations for this difference, e.g. differences in experimental treatments and sensitivity to physical disturbance among assemblages (Hansson et al., 2000).

Potential trawl disturbance effects in the study carried out in the Gullmarsfjord were also investigated on the basis of changes in temporal and spatial variability (Lindegarth et al., 2000a), which have been proposed to be more sensitive to detecting environmental impacts than tests based on the effects on means (Underwood, 1991; Chapman, Underwood and Skilleter, 1995). These analyses showed that trawling affected small-scale temporal and spatial variability in the structure of assemblages by counteracting the decreases in variability that occurred at the untrawled sites. The authors concluded that these changes in the overall structure of the assemblages of large macrofauna were relatively subtle compared with the changes caused by natural factors, because large temporal changes in benthos were demonstrated in both trawled and untrawled areas.

The Gulf of St Vincent in South Australia had not been trawled for at least ten to 15 years before the impacts of shrimp trawling were investigated by Drabsch, Tanner and Connell (2001). Three locations, 13 to 16 km apart were chosen, and each location included a control corridor and an adjacent trawled corridor that was trawled on average twice. The before-trawl sampling was carried out two months prior to trawling, and the post-trawl samples were collected within one week after trawling. Thus, the study used a replicate BACI design including three pairs of treatment and control sites. No consistent and unambiguous effects that could be attributed to trawling were detected in the statistical analyses. Large spatial and temporal variation was demonstrated, however. The authors concluded that the lack of an effect is most likely due to the light trawl gear and the low level of trawling intensity used in this experiment, which are characteristic of the fishing grounds in the area studied.

An area of coastal Maine (United States) that had been closed to shrimp trawling for 20 years was used to study the effects of trawling disturbance on a muddy (clayey-silt) bottom site (Sparks-McConkey and Watling, 2001). The study area of 1.82 by 2 km was covered by nine sampling stations, of which two were trawled four times. Macrofaunal samples (grab and core) were collected every three months for a year and a half prior to trawling and for six months after trawling. Immediately following the trawling disturbance, the total number of individuals, species richness and diversity in the trawled sites were significantly lower than in the control sites. Three months after trawling, these values for the trawled sites had returned to those of the control sites. Of the seven common polychaetes identified, four species decreased immediately after trawling and three species did not show any change in abundance. Three bivalves also showed a decrease in abundance immediately following the trawling disturbance. Nemertean abundances were significantly higher after trawling. Three months after trawling, the abundances of these common taxa, in general, were similar for trawled and control sites.

When interpreting these results it should be noted that most taxa showed large seasonal variations, and ecological parameters such as diversity and total abundance were significantly different between trawled and control sites at some stages before the trawling event. Accordingly, the most apparent feature of the multidimensional scaling analysis was changed species abundance of the common taxa, which is attributed to temporal variation and not trawling disturbance.

Two estuarine sounds in South Carolina (United States) were studied to evaluate the effects of commercial shrimp trawling on benthic infaunal assemblages (Van Dolah, Wendt and Levisen, 1991). In each sound, comparisons were made between an area closed to trawling and an area actively trawled during the fishing season. Grab samples were taken immediately prior to the trawling season and after five months of relatively intensive trawling. The number of species in both sounds showed a significant time effect, but no significant area effect, i.e. no difference between trawled and control areas. Species diversity, evenness and richness were similar in the trawled and control areas in one of the sounds both before and after trawling. In the other sound, species diversity increased in the control area between the two sampling periods, but not in the trawled area, which, however, still had higher diversity than the control area after five months of trawling. The total number of individuals differed considerably between control and trawled areas within each sound, between sounds and between sampling periods. Comparison of species composition and mean faunal abundance showed little evidence for trawl effects in either sound, but a significant decrease in the density of benthic infauna was observed between seasons. The authors concluded that the various community parameters assessed provided no clear evidence of trawl effects on the benthic infaunal communities. Most of the differences observed were attributed to natural seasonal variability.

Long-term impacts on the benthos were examined at two sites off the northeastern coast of England (United Kingdom), one at 80 m depth within a Nephrops fishing ground, and the other at 55 m located outside the main fishing area (Frid, Clark and Hall, 1999). Temporal changes of infaunal abundance and species composition were compared between the two sites over a 27-year sampling period, during which the intensity of fishing effort changed. Fluctuations in macrofaunal abundance at the site outside the main fishing ground reflected the abundance of phytoplankton, but this relationship broke down at the site within the fishing ground during a period of increasing fishing intensity. During this period of intense fishing activity, the total abundance of individuals of taxa likely to respond positively to fishing disturbance increased significantly and then subsequently declined when fishing decreased. Numbers of individuals in taxa likely to decline in response to fishing impacts did not change significantly between time periods with different levels of fishing effort. These two taxonomic groups did not vary at the site outside the fishing area.

The difference in the dynamics of these two sites, which were fished at different intensities, provides some evidence for the effects of fishing on the abundance and composition of coastal macrofauna. However, the sites studied also differed in several other aspects, e.g. depth, community structure and sediment type, which may have affected their responses to temporal changes in, for instance, phytoplankton productivity.

The whole Irish Sea has been intensively trawled for Norway lobster, and the only unfished sites are connected to wrecks. The short-term impacts of this trawl fishery were studied by comparing grab samples taken at two sites (at 35 and 75 m depth) before and 24 hours after experimental trawling (Ball, Fox and Munday, 2000). At the shallow site, most of the polychaetes species (scavengers or opportunistic species) increased in numbers following trawling, whereas most other species showed a decrease in numbers, although few comparisons were statistically significant. There were significant decreases in number of species, biomass, species richness and diversity after trawling. The fauna at the deeper site was too scarce to make the quantitative assessment of short-term effects possible.

The long-term effects were assessed by comparing the samples taken at the fished sites with samples taken at two nearby shipwrecks. The number of individuals and the biomass showed significant decreases between the wreck and the fished ground at the shallow site, and all parameters measured showed significant decreases between the wreck and the fished ground at the deeper site. Comparisons between trawling grounds and wreck sites were based on the assumption that they differed only in fishing intensity. As discussed in Chapter 2 Methodologies, however, wrecks are artificial reefs that may have several additional effects on the benthic community in their neighbourhood. Thus, clear evidence has not been provided for the conclusion made by this study's authors that the observed differences between the fauna of the two wreck sites and the nearby fished sites would appear to reflect genuine effects of fishing.

Hall et al. (1993) also used a wreck as their control site, and postulated that there is a gradient of increasing trawl disturbance with increasing distance from a ship wreck. They collected grab samples along transects running outwards (5 to 350 m) from a wreck site located in an area of the northern North Sea that has been heavily fished by otter trawl. The community variables studied (abundance, number of species, diversity) showed no correlation with distance from the wreck. Total numbers of individuals were, however, significantly correlated with sediment particle size. Thus, the spatial patterns in the benthic community that were demonstrated in this study did not appear to be consistent with an effect of fishing disturbance. The authors questioned whether studies of this kind, i.e. that use areas in close vicinity to ship wrecks as control sites, are adequate for testing the impacts of fishing disturbance. They stated that an important flaw in such studies is that there are no hard data on the distribution of disturbance, and it is simply hypothesized that a gradient exists.

The Mediterranean is markedly different from most other areas where impact studies have been conducted because of its oligotrophy, high level of salinity, high temperatures, negligible tidal currents and deep trawlable depths (Smith, Papadopoulou and Diliberto, 2000). As a result, this ecosystem may potentially be less robust and liable to disturbances. The impacts of a trawl fishery that takes place off Crete (Greece) for eight months and is then closed for four were studied by sampling at two sites in the trawled area and two control sites on either side of the trawl lane (Smith, Papadopoulou and Diliberto, 2000). Grab samples of macrofauna were taken periodically throughout the trawling and closed seasons. Towed-video operations demonstrated that trawling activity was confined to the trawl lane, and the sampling stations appeared to be consistent with the definition of trawled and control sites, although they where at different depths.

Total number of species was significantly higher on the control than on the trawled sites at some stage during the trawling season. This difference was mainly the result of higher numbers of species of echinoderms and sipunculids, which also showed greater abundances and biomass at the control sites. For crustaceans, molluscs and polychaetes the differences were rarely significant. Diversity was significantly higher and evenness significantly lower on the control than on the trawled sites. These ecological parameters showed a high degree of variability over time, and in most cases there were large differences between the two control sites, indicating temporal and spatial variability caused by factors other than trawling.

In another study conducted in the Mediterranean, experimental trawling was performed along two corridors that were swept entirely by the trawl once and twice, respectively (Sanchez et al., 2000). The infauna was sampled with a grab at 16 stations in the fished and adjacent unfished areas in order to evaluate the short-term effects. Samples were collected periodically for 150 hours after trawling on the site that was trawled once, and for 72 hours after trawling on the site that was trawled twice. There were no significant differences in the diversity indices (diversity, evenness and dominance) between the fished and control sites. The total numbers of individuals and species were not significantly different between fished and control sites during the first 102 hours after trawling, but increased significantly at the site trawled once compared with the control site 150 hours after the disturbance. For the area trawled twice, the total numbers of individuals and species changed significantly through time in both the fished and unfished sites. Based on these results, the authors suggested that sporadic episodes of trawling in muddy habitats may cause relatively few changes in community composition.

In a third Mediterranean study, the structure of the benthic fauna of two neighbouring gulfs of the Aegean Sea, one open and the other closed to trawling, were assessed in relation to natural and anthropogenic factors (Simboura et al., 1998). Although at the same depth and in close proximity, the two areas differed in sediment coarseness and organic content (higher percentage of sand and lower organic content in the trawling area).

The impacts of trawling have been indicated in some studies conducted on soft bottoms. However, evidence for clear and consistent changes attributable to trawling has not been provided from these experiments. The most prominent features of these studies are a lack of true and replicate control sites and pronounced temporal and spatial variability in community structures.

The area open to trawling had higher numbers of species (three times) and individuals (four to five times), and higher community diversity and species richness. Sediment type and organic content were regarded as the controlling factors for this differentiation of the two areas. Of the many parameters estimated in the study, only the higher number of polychaetes and the dominance of opportunistic polychaete species in the trawled area indicated a degree of disturbance that could possibly be linked to trawling. This study demonstrated considerable differences in the community structure of closely located benthic communities, which were attributed to natural parameters, and showed that trawling disturbance may be difficult to demonstrate owing to the masking of more dominant natural factors.

OTTER TRAWLING ON HARD BOTTOM HABITATS WITH ERECT STRUCTURES

Few studies of hard, rocky bottoms have been carried out, probably because sampling is difficult and expensive tools are required for visual observations. Freese et al. (1999) used video camera observations from a submersible to assess trawl impacts in the eastern Gulf of Alaska. A study area of hard bottom habitat (pebble, cobble, boulder) that had endured no or minimal trawling since the 1970s was identified by examining the trawling activity of the commercial fleet and by videotape recordings showing no evidence of trawling. This approach cannot confirm an undisturbed control site, but it is likely that the area had experienced low trawling activity. Shortly (from two hours to five days) after each trawl haul, the submersible made a transect along the trawl pass in the centre of marks made by the 60-cm-diameter tyre gear, which was a 5-m wide compact gear with no space between the tyres. Thus, in contrast to most other studies, all sampling was carried out in areas affected by the ground gear. A transect 50 to 70 m adjacent to each trawl path was used as a control site. Only easily visible and identifiable species were included in the analyses. These were 29 taxa greater than 5 cm in size.

The dominant substrate type was pebble (< 6.5 cm), and the trawl path was visible as a series of furrows caused by the tyre gear. The density of large sponges and anthozoans (corals) decreased significantly as a result of trawling. In the trawled transects, large proportions (55 to 67 percent) of these large erect sessile invertebrates were damaged. None of the motile invertebrates showed a significant reduction in density as a result of trawling, and only the ophiuroid Amphiophiura ponderosa was susceptible to damage (23 percent). In conclusion, this study demonstrated that boulders were displaced, and large emergent sessile epifauna were removed or damaged by the ground gear after a single trawl haul.

Considerable decreases in the abundance of large erect sessile invertebrates (e.g. sponges and corals) have been demonstrated to be a consequence of trawling. Habitats dominated by large sessile fauna may be severely affected by trawling disturbance.

In an investigation conducted on the continental shelf off northwestern Australia, attached benthos (> 20 cm) were counted from video transects, and comparisons were made among three sites, two that were trawled four times with demersal and semi-pelagic trawls, respectively, and one control site (Moran and Stephenson, 2000). The density of macrobenthos (mainly sponges, soft corals and gorgonians) decreased by 15.5 percent after each haul by the demersal trawl, leading to a reduction in density of about half after four trawl passes. The semi-pelagic trawl inflicted no detectable mortality on the benthos, but the catch rates of target fish species were low. Using effort data from logbooks, the annual mortality of macrobenthos was estimated to be less than 10 percent for most of the area fished by the commercial fleet.

The effects on sponges and corals of one path of a research trawl over a low-relief hard bottom habitat were studied in the South Atlantic Bight (Georgia, United States) by Van Dolah, Wendt and Nicholson (1987). The densities of individuals taller than 10 cm of three species of sponges and four species of corals (three octocorals and one stony coral) in the trawl path and in an adjacent control area were assessed by divers, and were compared before, immediately after and 12 months after trawling. The density of undamaged sponges showed a significant decrease immediately after trawling. Of the total number of sponges remaining in the trawled area, 32 percent were damaged. Most of the effected sponges were the barrel sponges Cliona spp., whereas the finger sponges (Haliclona oculata) and the vase sponges (Ircinia campana) were not significantly affected. Twelve months after trawling, the abundance of sponges had increased to pre-trawled densities or greater. The effects on the coral species appeared to be minimal in comparison with the sponges, as there were no significant differences between pre- and post-trawled densities for any of the species, although two species showed a non-significant decrease and some damaged corals were found immediately after trawling. The research trawl used in this study had a foot rope equipped with larger and more flexible rollers than that of the trawl used in the commercial shrimp fishery. Thus, supported by the findings of an unpublished study, the authors stated that commercial fishery is likely to cause more severe damage to the large sessile fauna of the area studied.

BEAM TRAWLING

Intensive disturbance by beam trawling has been shown to cause short-term changes in community structure through considerable reductions in abundance of infauna and epifauna. The long-term effects have not been studied.

Most studies on the impacts of beam trawling have been conducted in the North Sea and the Irish Sea, where some areas have been intensively trawled for many decades. As hardly any areas in the North Sea have not been affected by commercial trawling (except for those in the vicinity of wrecks), few if any areas are suitable as control sites. Most studies have therefore assessed trawling impacts by comparing samples taken before and immediately after experimental trawling on commercial fishing grounds.

One study applying this approach was conducted in the southern North Sea to determine the impacts of a heavy 12-m beam trawl (7 000 kg) (Bergman and Hup, 1992). A small quadrant was trawled three times, and samples were taken before trawling started, after the first and the third trawling events and again two weeks later. The direct effect of this disturbance was a significant decrease in abundance (40 to 60 percent) of one echinoderm species and two species of polychaete, whereas the polychaete Magelona papillicornis increased in density (35 percent). No effects were found on the densities of Ophiura sp., molluscs (about ten species) or other species of worms (about 20 species). The decreased numbers of some species were explained by destruction during the passage of the trawl and removal by the trawl net. As the bycatch was returned to the sea after sorting on board, the effect on the benthic community depended on the survival rate of these species. The immediate effects demonstrated for some species in this study appeared to be considerable, but the authors stated that such direct effects cannot be extrapolated to the long-term effects of beam trawling on the benthic community.

In another North Sea experiment, the direct mortality of mega- and macrofauna caused by different beam trawls (12 m and 4 m wide, with tickler chains with chain mats) was also determined by the differences in densities before and after experimental trawling (Bergman and van Santbrink, 2000). The experimental corridor was trawled at an average frequency of 1.5 tows, but the mortality estimates were converted to those for a single passage of the trawl. Several taxa did not show statistically significant mortalities, but those that did were considerable. Small-sized bivalves and crustaceans showed direct mortalities of up to 22 percent, and annelid worms of up to 31 percent. In megafaunal crustaceans and bivalves, mortalities of up to 49 and 68 percent, respectively, were found. The majority of the species studied showed similar mortality caused by 4 m and 12 m beam trawls in silty and sandy areas, but there was a tendency for lower mortalities from trawls with a chain mat compared with those with tickler chains. Mortality in some infaunal species was higher in silty than in sandy areas, and small species tended in general to show low mortalities compared with larger species. The estimated direct mortality incorporated animals caught by the trawl and animals damaged or exposed in the trawl track, of which the former were insignificant owing to low catch efficiency for invertebrates. Based on different assumptions (e.g. about spatial species distribution and commercial trawling frequency), annual fishing-induced mortality in megafaunal populations was estimated to range from 5 to 39 percent on the Netherlands continental shelf, and was mainly due to the 12 m beam trawl as this is the dominant gear type, with an average trawl frequency coverage (i.e. number of tows on a given site of the fishing ground) of 1.23 (in 1994).

The Irish Sea is a heavily trawled area, particularly the northern part (Kaiser et al., 1996). The effects of beam trawling on benthic infauna were determined in two costal communities in the northeastern Irish Sea, one characterized by megaripples and mobile sediments and the other by featureless and stable sediments (Kaiser and Spencer, 1996). Three corridors were fished ten or 20 times with a 4 m commercial beam trawl (3 500 kg) fitted with a chain matrix. These and adjacent unfished control areas were sampled 12 hours after trawling. The mean total number of species and the abundance of individuals were 2.4 and 5.7 times higher, respectively, in the area with stable than in the area with mobile sediments. The community of the former was affected by the trawling disturbance, with the number, abundance and diversity of taxa significantly lower in the fished than the unfished areas. The abundance of nine of the 20 most common taxa was significantly lower in the fished areas, by as much as 50 percent for the two most abundant taxa (Urothoe and Ampelisca). In the area with mobile sediments, there were no significant differences in species numbers, abundance or diversity between the fished and unfished sites. Thus, this study demonstrated the short-term effects of beam trawling on infauna taxa that live in stable sediments, but similar effects were not detected on animals inhabiting mobile sediments that are subjected to frequent natural disturbances.

As part of the same study, the effects on the megafaunal component (> 10 mm) of the community were studied immediately (approximately 24 hours) after trawling and six months later (Kaiser et al., 1998). The two areas with stable and mobile sediments also differed significantly with regard to megafaunal community variables (number of species, number of individuals and diversity). As with the infauna, only the megafaunal samples taken from the area with stable sediments immediately after fishing revealed significant differences between the fished and control areas. A reduction in abundance of the polychaetes Aphtodita aculeate and Nephtys spp. contributed most of this dissimilarity, but changes in number of species, number of individuals or diversity as a result of fishing disturbance were not detected. The differences were no longer apparent six months after the trawling disturbance. However, there were marked seasonal changes in the community structure. In conclusion, this part of the study revealed that only subtle changes in community structure were caused by trawling, whereas the effects caused by seasonal fluctuations and natural disturbances were more pronounced.

Most studies have determined the impacts on the benthic community structure by recording changes in the abundance and biomass of different species, but Jennings et al. (2001) also investigated the effects of trawling on the trophic structure of the community studied. Trawling effort at the sites studied was determined from records of vessel sightings by fishery protection aircraft, and the authors assumed that the number of trawlers sighted per unit of searching effort was linearly proportional to the trawling disturbance. This method does not give an accurate indication of the actual level of disturbance at the site where the invertebrates were sampled, because the samples were collected at a much finer spatial scale than that of the overflight data. Two areas in the North Sea subjected to different levels of trawling (a threefold difference) were investigated. Each area was divided into sites of 5 by 6 nautical miles, and within the two areas there were 27- and tenfold ranges, respectively, in the trawling disturbance among the sites.

The total biomass of infauna and epifauna in the most intensively trawled area decreased significantly with trawling disturbance. This decrease was most pronounced for the biomass of infauna, which showed an order of magnitude decrease when trawling disturbance increased 27-fold and was due to decreases in the biomass of bivalves and spatangoids, while there was no change in polychaetes. At the less intensively trawled area there were no significant trends in total biomass in relation to fishing disturbance, but there was a significant increase in polychaetes. No changes in the trophic structure of the community related to trawling disturbance were found. In conclusion, the study suggested that highly intensive trawling disturbance led reduced biomass of infauna and epifauna and dramatic changes in the composition of infauna, but these changes were not reflected in the mean tropic level of the community.

SCALLOP DREDGING

A comprehensive study conducted on a spatial scale that is relevant to commercial dredging demonstrated a decrease in number of species and considerable reductions in abundance of several species, but dredging-induced reductions in density were small compared with temporal and spatial changes.

A comprehensive study on scallop dredging was conducted in Port Phillip Bay (Australia) on a spatial scale that is relevant to commercial fishery (Currie and Parry, 1996). A large area (600 x 600 m) was dredged over three days, until the entire site had been passed over an average of twice by the dredge. The abundances of infauna at this and an adjacent control site were determined from grab samples taken on three sampling dates before and six after (a few hours, three weeks, and three and a half, five, eight and 14 months) the experimental dredging. The dredging caused a significant decrease in the number of species on the dredge site compared with the control site (10 percent significance level was used). This difference persisted for eight months, but after 14 months the number of species was similar on both sites. The total number of individuals showed large seasonal changes, and there were no differences between the dredged and the control sites. Bray-Curtis dissimilarity measures between the sites increased significantly after dredging and persisted for 14 months, i.e. the duration of the study. In addition, the multidimensional scaling (MDS) ordination showed increased differences between the two sites after dredging, but changes caused by seasonal and interannual variations were more pronounced.

Of the ten most abundant species, six showed a significant decrease in abundance (28 to 79 percent) and one species increased (141 percent) after dredging. The duration of these impacts varied among species, but few lasted beyond eight months after dredging. In this study, a pronounced ecological gradient was detected between the eastern and western parts of both the dredged and the control sites. This difference was seen in the distribution of many species, and the impacts demonstrated were not consistent between the eastern and western parts. Although impacts of scallop dredging were demonstrated in this large-scale study, the spatial heterogeneity - together with seasonal and interannual changes - makes interpretation more difficult. The authors stated that the reductions in density caused by dredging were usually small compared with annual changes in population density.

Currie and Parry (1999) report a similar study involving two additional areas in Port Phillip Bay with different soft substrates, in which the impacts on epifauna were also investigated. The dredge caught mostly scallops, with the bycatch of other epibiota being typically less than 5 percent. Damage to bycatch species was low except for spider crabs (up to 55 percent). In one of the areas, three of the ten most abundant infaunal species decreased significantly in abundance by 20 to 40 percent. Changes to community structure caused by dredging were small compared with differences between the areas studied. The authors suggested that the relatively small effects of dredging demonstrated in these two studies may be related to the low abundance of epifauna other than scallops in Port Phillip Bay.

Two experimental sites were selected for a study in New Zealand, one in an area exploited by commercial scallop fishers and the other in an unexploited area (Thrush et al., 1995). A scallop dredge (2.4 m wide, with 10-cm-long teeth) was towed through half of each study site (five parallel tows) to create a dredged site and an adjacent control site. Core samples were collected by divers within two hours and again three months after dredging. The authors used a 10 percent significance level, as in their view Type II error is at least as important as Type I error when documenting impacts on the environment. In the unexploited area, the total numbers of individuals and species were significantly lower in the dredge site than the control site immediately after dredging. Reduced densities were observed for four crustacean species, three polychaetes and one bivalve. Some of these species also showed lower densities in the dredged site three months after dredging, and three species showed significant temporal changes in density during this period. In the exploited area, five species showed significant density changes over time, and patterns of density change between the dredged and the control site were less clear. Only two bivalves and one crustacean species showed consistently lower densities in the dredged site on both sampling occasions, and four species showed a marked increase in density in the dredged site after three months. The total numbers of individuals and species decreased as an effect of dredging. This experimental assessment, which was quite conservative (i.e. low intensity of disturbance in a small area), demonstrated that the macrobenthic community structure in dredged areas differed from that in undredged areas for at least three months.

Rapido trawls resemble a toothed beam trawl and are used around the coast of Italy to catch sole and scallop (see diagram and description in Hall-Spencer et al., 1999). They are lightly built (3 m wide, 170 kg), and are towed at high speed (10 to 13 km/hour). As the rapido trawl has teeth and no tickler chains or chain matrix, its effects on the benthic habitat are more likely to resemble those of a scallop dredge. The impact of this gear on a commercial scallop ground in the Adriatic Sea (Mediterranean) was studied by making seven tows along a 60-m wide corridor (Hall-Spencer et al., 1999). A video sledge was towed along the corridor prior to trawling and again 15 hours after trawling, and the densities of visible organisms were determined from the video recordings. Significant reductions in abundance (> 70 percent) on the trawled track were found for several components of the macrofauna, e.g. the large bivalve Atrina fragilis, the sea cucumber Holothuria forskali, the sea anemone Cerianthus membranaceus, tunicates and naticid egg coils. Observations during trawling revealed that most organisms in the path of the trawl passed under or through the net, and analysis of the catch showed low total bycatch biomass (19 percent), which was dominated by tunicates, echinoderms and molluscs. Lethal mechanical damage of the taxa caught varied from < 10 percent in small, resilient taxa such as hermit crabs and gastropods to > 50 percent in large, fragile organisms such as A. fragilis and tunicates. The short-term effects demonstrated in this study included reductions in the structural complexity of the habitat and the selective removal of a large proportion of the benthos.

To obtain samples from an undisturbed area, experimental trawling with a rapido trawl was conducted near a wreck (Pranovi et al., 2000). The disturbed line was trawled only once, and sampling by diving ensured that samples were taken exactly within the trawl track and at an adjacent control site. The track was sampled immediately after trawling and seven days later. Samples were also collected at a nearby fishing ground. The findings of this study cannot easily be interpreted, as there are several discrepancies between the text and the tables regarding whether differences were significant or not. In general, however, common macrofaunal taxa decreased in density immediately after trawling, but increased again after one week and were often higher than they were in the controls. In the comparison between the control area and the neighbouring fishing ground, biomass, numbers of individuals and taxa and numbers of many taxa were significantly lower at the fishing ground. The community in the exploited area was also less diverse and dominated by only three phyla (Mollusca, Arthropoda and Echinodermata). Analysis of the meiofauna showed that some taxa decreased after trawling while others increased. Compared with the control area, the fishing ground showed significantly higher numbers of individuals, but lower evenness. Thus, while the immediate effect of rapido trawling (one haul) was negative, the number of individuals and the total number of taxa increased after one week. These species were mainly scavengers.

Numerous studies have assessed the impacts of scallop dredging and, although several have important limitations, they have all indicated effects of varying degrees. Short-term reductions in species abundance and decreases in number of individuals were the most prominent features.

The area around the Isle of Man in the Irish Sea has been heavily fished by scallop dredges. Fishing effort data from logbooks recording effort per 5 x 5 nautical mile boxes were used to compare five areas subjected to low and five areas subjected to high dredging intensity (Kaiser et al., 2000). Infaunal samples were collected using an anchor dredge, and epifauna were sampled by a 2-m wide beam trawl. Both the total biomass and the abundance of epifauna differed between the areas of low and high fishing intensity, whereas the infauna samples differed in biomass only. However, the direction of these differences is not given in the paper. There were no significant differences in number of species or diversity indices. Abundance-biomass curves (ABC) indicated that the heavily fished areas were dominated by higher abundances of smaller-bodied organisms, whereas the less intensively fished areas were dominated by fewer, larger-bodied biota. Although this study indicated that scallop dredging has led to changes in community structure, it must be noted that the study had several deficiencies, e.g. the areas compared were all disturbed, and the sampling devices used were not strictly quantitative. Furthermore, these areas were located far apart and differed in habitat type (depth and sediment type), which was shown to affect community structure. Despite these shortcomings, the authors generalized their results and stated that "... any of the large, bottom-fishing gears such as rock-hopper otter trawls and beam trawls will have similar effects".

A comparison between sites subjected to low (closed to towed gear), medium (seasonally open to towed gear) and high (open to towed gear) levels of fishing effort with scallop dredge, beam trawl and otter trawl was made on fishing grounds along the southern coast of the United Kingdom (Kaiser, Spencer and Hart, 2000). Information on the type of towed gear used or precise measures of fishing effort at the sites sampled were not available to the authors. Epifauna was sampled with a 2-m beam trawl, and infauna with an anchor dredge. Several environmental parameters (depth, grain size, mass of stones and broken shell, RoxAnn E1 and E2 values) varied significantly among the sites studied. These habitat differences had a more apparent effect on community structures (total number of individuals, number of species, diversity) than did level of fishing effort. The biomass data revealed significant differences among areas of high, medium and low fishing effort. There was a general decrease in less mobile, larger-bodied and fragile fauna and an increase in the more resilient, mobile fauna as fishing disturbance increased. Although the responses of individual taxa to fishing disturbance were not consistent for the different habitat types, the biomass of sessile fauna such as soft corals and hydroids was higher in the areas closed to towed fishing gear. The results of this study indicate the effects of trawling on benthic community structure, but this study has several limitations (e.g. lack of fishing effort data and appropriate control sites), and the habitat differences between the sites studied were shown to have the most apparent effect on community structures.

Thrush et al. (1998) counted large epifauna (video transects) and sampled macrofauna (grab or suction-dredge and core sampling) at 18 sites that were ranked in terms of fishing pressure from trawling, Danish seining and scallop dredging. The ranking of fishing pressure was based on fisheries legislation (e.g. closed areas) and information from fisheries managers. Dredging was considered to cause a stronger impact than trawling or seining. Such ranking should be regarded as rough, imprecise and, to some extent, subjective, so the results of the study must be interpreted with great caution. Based on results from small-scale experimental disturbance studies, a priori predictions about the responses of benthic communities to trawl disturbances were made and tested at larger scales. Fishing-pressure rank was found to account for 15 to 20 percent of the variability in community structure. Of the ten predictions tested with the core samples, fishing pressure was shown to affect significantly the density of echinoderms, polychaete/mollusc ratios (but in ways that were contrary to the prediction), the diversity index and the number of species. However, using the grab/dredge samples, fishing pressure was not a factor that affected these ecological parameters. The results from the grab/dredge samples showed significant effects on only deposit feeders, but these too were contrary to the prediction.

In this study, decreasing fishing pressure was seen to change the benthic communities in the direction predicted. However, few of the ecological parameters tested were significant, and the findings from the core samples were not supported by those from the grab/dredge samples. Given the lack of appropriate reference sites, this analysis used fishing intensity ranking based on legislation and information from fisheries managers that identified areas of different exploitation rates. One problem of this approach is that it does not identify any unambiguous cause-and-effect relationship. The study was integrated over a variety of habitat types, and the differences in community structure that were demonstrated for areas of different exploitation rate may not be a response to fishing disturbance but simply the result of fishers avoiding certain areas (e.g. because of reefs and other obstacles) and preferring other zones with different community structures.

Georges Bank (northwest Atlantic) has been dredged for scallops for decades, and the impacts on the benthic megafauna on a gravel habitat have been studied by Collie, Escanero and Valentine (1997). Two disturbed and six undisturbed sites were identified based on a combination of side-scan sonar recordings, video observations and records of dredging effort (available only for two of the six sites). Samples of benthic megafauna were obtained with a dredge and standardized per unit of sediment collected. The abundance of organisms, biomass and species diversity were significantly greater at undisturbed than disturbed sites, whereas evenness was significantly greater at disturbed sites. Although the data did not allow statistical testing of the species richness measure, they indicated that species richness was higher at undisturbed than disturbed sites. The authors stated that their study has several limitations and must be interpreted with caution. The sampling dredge was considered to be semi-quantitative, its penetration into the gravel was not constant and the estimates of the area sampled were imprecise. The benthic samples were therefore not strictly quantitative and comparable.

In this investigation, video recordings and still photographs were also taken along transects within each site (Collie, Escanero and Valentine, 2000). The still photographs showed that the percentage cover of the colonial polychaete Filograna implexa was significantly greater at the undisturbed sites. However, the percentage cover of low-encrusting bryozoans was significantly higher in the disturbed sites. The percentage cover of plant-like animals (bushy hydroids and bryozoans) was greater at deep undisturbed sites, but the effect of dredging was reversed at the shallow sites where the disturbed site had higher percentage cover of these organisms. The same contradictory results between deep and shallow sites were seen in the percentage cover of emergent epifauna estimated from the videos. Similarly, there were no consistent differences between disturbance levels in terms of numerical abundance of megafauna, species diversity or evenness. Analyses of the abundance of organisms that could be identified in the videos and still photographs showed that most of these taxa were more abundant at the undisturbed sites.

Although dredging disturbance is a likely explanation for the differences in community structure demonstrated at the studied sites, the approach used in this study has some serious shortcomings. Fishing effort data (on a large scale) were available for only two of the sites, and side-scan recording is a rough and descriptive way of determining sediment type and degree of bottom disturbance. The sea bed was surveyed by video transects to identify stations for benthic sampling, and sites where epifauna were absent or present were used as disturbed and undisturbed sites, respectively. However, the presence or absence of epifauna may have other causes than dredging, e.g. differences in sediment characteristics and, thus, habitat type. In fact, the sediment distributions were patchy and differed between the disturbed and the undisturbed sites. Furthermore, the abundance of sea scallop (the target of commercial dredging) was five times higher at the disturbed (dredged) than the undisturbed sites (Collie, Escanero and Valentine, 2000; Table 3A). This difference (and possibly other changes demonstrated in the study) cannot be attributed to dredging disturbances, but simply reflects fishers' preference for areas with high densities of the target species, and thus indicates that the sites studied involved different habitat types.

The problem of finding suitable control sites among the scallop grounds in the Irish Sea was overcome by using samples collected before the initiation of intensive fishing (Hill et al., 1999). Samples of epifauna and infauna collected from three sites over the period 1946 to 1952 were compared with recent samples taken from the same sites. Since the 1960s, two of these sites have been heavily dredged, while the third site has been relatively unaffected by dredging. At the heavily fished sites, the number of species, species richness and diversity were significantly greater in the recent samples compared with the historical samples, whereas the index of Simpson's dominance was significantly greater in the historical samples. Species composition was clearly different between recent and historical samples. This difference included the presence of some larger and fragile species (two echinoids and one bivalve) in the historical samples, but not in the recent samples, and a higher polychaete to mollusc ratio in the recent samples. Significant changes were also found at the site unaffected by dredging. The number of individuals and the index of Simpson's dominance decreased, and evenness increased between the historical and recent samples. The species composition also changed over this time period, and the polychaete to mollusc ratio increased. The authors believe that this study has a serious shortcoming because different types of sampling gear were used to collect the historical and recent samples. Natural long-term changes might have occurred over this time period because the community structure also changed at the undisturbed site. Furthermore, some of the changes found at the heavily fished sites were the opposite of those expected to result from physical disturbance. Although this study showed that community structures had changed since 1950, the cause of the change was not clearly demonstrated.

Changes attributable to dredging were not observed in areas exposed to natural stress, e.g. wave action, eutrophication and salinity fluctuation.

The effects caused by a modified scallop dredge (1.2 m wide, 12-cm-long teeth separated by 8 cm, with the chain net removed) were examined in a shallow Scottish loch that is exposed to wave and tidal actions (Eleftheriou and Robertson, 1992). The dredge was towed a number of times over the same track, and after 0, two, four, 12 and 25 tows sampling (grab and diver observations) was conducted in the middle of the track and in one small control area. No significant changes in abundance, biomass, diversity or evenness of infaunal species sampled by grab that were attributable to the dredging were observed. Divers observed some damage and mortality to organisms such as Echinocardium, Asterias and Ensis, and high mortality of sand eel (Ammodytes) was observed.

Studies were made in an area of the Limfjord (Denmark) that was closed to mussel dredging ten years prior to the experiment (Hoffmann and Dolmer, 2000). The epifauna at two sites within and two sites outside the closed area was sampled by divers. The species composition differed significantly among the four sites, but these spatial differences could not be attributed to dredging disturbance. The authors suggested that other factors, such as oxygen depletion, have a much greater impact on the spatial variability in this ecosystem. This ecosystem is also subjected to large fluctuations in salinity, temperature and eutrophication.

TABLE 3
Studies to investigate the impacts of beam trawling on benthic habitats and communities

Study site

Depth
(m)

Bottom type

Dominant species

Sampling techniques

References

North Sea

Ca. 30

Fine/medium sand

Crustacean, mollusc, echinoderm, polychaete

Grab, boxcorer, beam trawl

Bergman & Hup, 1992

North Sea

<30-50

Silt, sand

Mollusc, echinoderm, crustacean, polychaete

Grab, boxcorer, dredge

Bergman & Santbrink, 2000

North Sea

40-60 and 60-80

Sand, muddy sand

Bivalve, spatangoid, polychaete

Beam trawl, dredge

Jennings et al., 2001

Irish Sea

26-34

Medium/coarse sand, gravel

Amphipod, polychaete, echinoderm

Grab, dredge, side-scan, RoxAnn

Kaiser & Spencer, 1996; Kaiser et al., 1998

TABLE 4
Studies to investigate the impacts of scallop dredging on benthic habitats and communities

Gear

Study site

Depth
(m)

Bottom type

Dominant species

Sampling techniques

References

Scallop dredge

Australia

12-16

Sand, silt

Crustacean, polychaete, mollusc

Grab, sledge, video, divers

Currie & Parry, 1996; 1999

Scallop dredge

New Zealand

24

Coarse sand

Polychaete, crustacean, mollusc

Core

Thrush et al., 1995

Rapido trawl

Adriatic Sea

25

Sand

Bivalve, ophiuroid, sponge, tunicate, holothurian

Video

Hall-Spencer et al., 1999

Rapido trawl

Adriatic Sea

24

Sand

Mollusc, polychaete, crustacean

Side-scan, divers, water-lift sampler, core

Pranovi et al., 2000

Scallop dredge

Scottish bay

6

Sand

Polychaete, crustacean, mollusc, echinoderm

Grab, core, divers, photos

Eleftheriou & Robertson, 1992

Mussel dredge

Danish fjord

2-7

Not given

Porifera, anthozoa, mollusc, echinoderm

Divers

Hoffmann & Dolmer, 2000

Scallop dredge

Irish Sea

Not given

Coarse sand, gravel

Soft coral, echinoderm, mollusc

Dredge, beam trawl

Kaiser et al., 2000

Scallop dredge, beam and otter trawl

Coast of Devon, UK

15-17 and 53-70

Fine, coarse-medium sand

Soft coral, hydroid, echinoderm, crustacean, mollusc

Dredge, beam trawl

Kaiser, Spencer & Hart, 2000

Trawl, seine, dredge

Hauraki Gulf, NZ

14-35

Clay, silt, shell fragments

Not given

Side-scan, camera, grab, dredge, core

Thrush et al., 1998

Scallop dredge, otter trawl

Georges Bank

42-90

Pebble, cobble

Mollusc, polychaete, crustacean, echinoderm

Dredge, video, photos

Collie, Escanero & Valentine, 1997; 2000

Scallop dredge

Irish Sea

Not given

Not given

Polychaete, mollusc, echinoderm, crustacean

Dredge, grab

Hill et al., 1999


Previous Page Top of Page Next Page