By Franz J. Mueter
School of Resource and Environmental Management
Simon Fraser University
8888 University Drive Burnaby, British Columbia, CANADA, V5A 1S6
Phone: (604) 291-3491, FAX: (604) 291-4968
The National Marine Fisheries Service (NMFS) is currently undertaking a comprehensive review of Fisheries Management Plans for the Gulf of Alaska (GOA) and Bering Sea/Aleutian Island regions. Findings to date are summarized in the Draft Programmatic Supplemental Environmental Impact Statement (PSEIS). The Marine Conservation Alliance (MCA) has hired me to conduct a review of the scientific baseline used in the PSEIS to describe the effects of fishing gear on benthic fish habitat. The MCA is a non-profit entity formed by Alaska and North Pacific coastal residents, communities, fishermen, fishing organizations, support industries, western Alaska Community Development Quota (CDQ) organizations, and others who are directly and indirectly involved in the North Pacific (Alaska) groundfish fisheries (www.marineconservationalliance.org). At the request of the MCA, I am submitting my review as public comment on the draft PSEIS as an attachment to the MCA’s more extensive public comment on the PSEIS.
Section 3.2 of the PSEIS includes a review of the effects of fishing gear on benthic habitat and communities. One of the policy alternatives evaluated as part of the PSEIS (Alternative 5) examines a management policy framework that places greater emphasis on objectives to protect, conserve, and restore marine habitat, while providing for sustainable groundfish fisheries. This alternative focuses on bottom trawl fisheries for flatfish, Pacific cod, and rockfish because of concerns over adverse effects of these fisheries on essential fish habitat (EFH).
The expected benefits to the measures in Alternative 5 that would further restrict trawling in the fisheries off Alaska are based on a review of trawling patterns off Alaska, a summary of the trawling impact literature from shelf areas around the world, and a model of the effects of the proposed measures on bycatch of HAPC biota (PSEIS, Section 3.2.1). Fishing effort in the Bering Sea was indexed on a 100 km2 grid, indicating that fishing effort at this scale is highly concentrated in specific areas, while the majority of the Bering Sea experienced low or very low trawl intensities (PSEIS, Fig. 3.2-1). Fishing effort in the GOA was quantified (number of trawls per km2) on a 25 km2 grid, suggesting a similar concentration of fishing effort in certain areas, while large areas of the shelf experienced low density trawling on average from 1990-1998 (PSEIS, Fig 3.2-3, 3.2-4).
Compared to other shelf areas, there has been a notable lack of studies on the impacts of trawling in the Northeast Pacific. Two relevant studies are summarized in the PSEIS. McConnaughey et al. (2000) examined the impacts of bottom trawling in a shallow, soft-bottom area of the Bering Sea and reported higher densities of some sedentary macrofauna in historically unfished areas, differences (both positive and negative) in the abundance of several macrobenthic species between heavily fished and unfished areas, and a greater diversity of macrofauna in unfished areas. The only study in the GOA to date examined the immediate effects of single passes with a commercial trawl on benthos (Freese et al. 1999). Direct observations indicated that 14 – 67% of large sessile epifauna was damaged by the trawl and densities of these epifaunal organisms were significantly higher in unfished reference sites. Neither damage to nor a reduction in densities of motile invertebrates was detected (Freese et al. 1999).
In addition to these case studies, general observations of trawling impacts on benthic habitat were summarized based on a recent review (Auster and Langton 1999) and other selected references. The PSEIS adopts the conclusion of Auster and Langton (1999) that trawl impact studies generally show similar classes of impacts. In particular, the PSEIS summarizes evidence for changes in physical substrate characteristics, as well as short- and long-term changes in benthic community composition. The PSEIS notes the lack of data on the effects of fixed gear (longline and pot gear) on habitat. Therefore, it concludes that an assessment of the relative effects of fixed gear and mobile gear on different habitats is currently not possible.
Nevertheless, a review of these trawl impact studies appears to provide the main justification for concluding that allocations of catch to fixed gear will reduce fishing impacts on habitat (PSEIS, p. 4.1-119). As well, gear allocation and gear restrictions were ranked among the best tools for achieving the policy objective of Alternative 5, i.e. to protect and restore essential fish habitat (PSEIS, Table 4.1-28). Therefore, the PSEIS recommends management actions for protecting habitat (Alternative 5) that have a strong focus on re-allocating fisheries to pelagic trawls and fixed gear, where possible (PSEIS, Table 4.1-20). Furthermore, based on Auster and Langton (1999) and McConnaughey et al. (2000), reductions in bottom trawling were judged to substantially improve benthic biodiversity off Alaska.
Given the substantial impact of Alternative 5 on the trawl and fixed gear fisheries, it is important to carefully evaluate the basis for the conclusions of the PSEIS. In particular, the limited number and scope of studies in Alaska warrant an evaluation of other relevant literature to understand potential effects of trawling and fixed gear on all aspects of the benthic ecosystem. While the PSEIS makes mention of a limited number of alternative trawl impact studies, it relies primarily on the review by Auster and Langton (1999). However, a large number of reviews and case studies on the effects of fishing on habitat have appeared in the literature in recent years and have added to our current understanding of the effects of trawling on habitat (e.g. Collie et al. 2000, Eleftheriou 2000, Kaiser and de Groot 2000, Moore and Jennings 2000). Hence, it is timely to review recent studies and other relevant work not included in Auster and Langton (1999) or in Section 3.2 of the PSEIS.
This document therefore critically reviews the evidence for adverse impacts of mobile and fixed fishing gears on benthic habitat and communities as cited in Auster and Langton (1999). In addition, I surveyed more recent literature on the subject, briefly summarize new findings, and evaluate whether findings can be generalized to the Northeast Pacific. Where appropriate, reference will be made to the PSEIS.
Summary of Findings
Auster and Langton (1999) divide impacts of fishing gear into effects on substrate (structural components of habitat), short-term effects on benthic communities, and long-term effects on benthic communities. I adopted the same distinction among potential impacts in this review.
Impacts on Substrate (including sessile epifauna)
Reported impacts of mobile fishing gear on living and non-living substrate include:
- overturning boulders (e.g. Freese et al. 1999);
- smoothing and compacting soft sediments (e.g. Schwinghamer et al. 1998);
- resuspension of sediments (Churchill 1989, Fonteyne 2000);
- removing and/or damaging sessile epifauna such as sponges, sea pens, corals, hydroids, and eelgrass (e.g. Collie et al. 1997, Freese et al. 1999, Pitcher et al. 2000).
The physical impact on non-living substrate is generally thought to be minor, particularly for lighter gear such as otter trawls, whose impact is largely restricted to the doors (Hall 1999). Two recent studies have examined effects of trawling on sediment characteristics. Schwinghamer et al. (1998) examined the effects of trawling on sediments of a sandy bottom ecosystem on the Grand Banks at a depth of 120-146 m. They fished otter trawls with trawl gear similar to the gear used in Alaska at three sites and at high intensities (12 trawls within 5 days). The physical effects of otter trawling at this site were moderate and recovery occurred within a year (Schwinghamer et al. 1998). Fonteyne (2000) examined the effects of a beam trawl on sandy substrate at 20-30 m depth. Tracks completely faded after 52 h and resuspended material (fine sand fraction) settled down within a few hours. Clearly, measurable effects on non-living substrate do occur, but the ecological implications are not known and may be positive or negative. While the displacement of large cobble and boulders may disrupt habitat features (Auster and Langton 1999, Freese et al. 1999), sediment turnover can oxygenate deeper layers and release buried organic matter or nutrients, which may increase benthic productivity. For example, delayed diatom blooms have been observed in trawl door tracks in shallow habitat, possibly due to the release of nutrients (Brylinski et al. 1994). Conversely, resuspended material is likely to reduce light levels and temporarily decrease benthic productivity. Generally, physical impacts of trawling on soft-bottom, high-energy areas such as the shallow parts of the Bering Sea (Marlow et al. 1999) are relatively minor compared to natural disturbance from wind mixing or tidal currents (Churchill 1989, DeAlteris et al. 1999, Hall 1999).
There is general agreement in the literature that trawl gear can remove and/or damage living structural components in areas where the substrate consists of large attached epifauna. This includes the HAPC fauna for which bycatch rates of different gear types were summarized in the PSEIS. Vulnerable habitats include sponge-coral hard bottom, bryozoan beds, mussel beds, and mixed substrates with attached epifauna (for examples, see reviews in Jennings 1998, Auster and Langton 1999, Kaiser and de Groot 2000, Moore and Jennings 2000). One study has been conducted on vulnerable habitat in the eastern GOA (Freese et al. 1999). However the latter study took place in an area that is currently closed to trawling. The shelf in the eastern GOA differs considerably from other shelf areas in the GOA in terms of topography, sediment characteristics, and benthic community composition (Hampton et al. 1986, Mueter 1999). Therefore, results may not be applicable to other areas in the GOA with different epibenthic communities. Unfortunately, very little is known about the substrate characteristics of different areas in Alaskan waters. While a number of areas of particular concern have been identified to date and have been closed to trawling (PSEIS, Table 4.7-2), it is equally important to identify areas where effects of trawling are expected to be minor or where recovery might be expected to be fast.
There is little information on recovery rates of attached epifauna, but they are likely to vary substantially among sites. Collie et al. (1997) found that a gravel site that had been dredged for scallops and was subsequently closed to fishing showed signs of recovery after 2 years. In contrast, Kaiser et al. (1998) found that epifaunal communities that had been trawled experimentally with a beam trawl in relatively shallow water (35 m) were indistinguishable from controlled unfished areas after only 6 months. Similarly, after a single trawl pass, sponges and corals on a hard-bottom substrate off Georgia fully recovered to pre-trawling levels within a year (Van Dolah et al. 1987). Thus fishing intensity and frequency are important considerations in evaluating long-term effects of trawling on the substrate. Some areas that are vulnerable to short-term effects of fishing appear to have relatively short recovery times and may be able to tolerate fishing at low intensities.
In heavily fished areas, the removal of large epibenthic organisms can lead to a long-term reduction in structural complexity and declines in the abundance of fishes associated with the epibenthic community (Jennings 1998). At present, evidence for such long-term indirect effects are largely from tropical reef systems (Jennings 1998). For temperate shelf systems, the wider ecological consequences for fish and invertebrate species are poorly understood. The main concern is over structurally complex substrates that provide important habitat for juvenile fishes and invertebrates. However, unlike the direct effects of the removal of target and bycatch species, effects of fishing on habitat have rarely been associated with long-term changes in fish populations (see below). Exceptions are Sainsbury et al. (1997), who showed that fish productivity along the Northwest continental shelf in Australia correlated well with the loss of attached epifauna, and Kaiser et al. (2000), who studied chronic fishing effects on two soft-sediment communities. The latter study indicated that chronic fishing caused a shift from communities dominated by large, sessile or attached epifauna to communities dominated by small, infaunal species in parts of the North Sea.
In summary, effects of trawling on physical sediment characteristics are expected to be minor, but effects on biogenic substrates, particularly those dominated by large sessile epifauna, can be substantial. However, recovery rates and the impact of reductions in sessile epifauna on other species are poorly understood. To understand potential impacts of both mobile and fixed gear on the structure of benthic habitats in the Gulf of Alaska and Bering Sea, the distribution of different substrate types, as well as the small-scale distribution of fishing effort, need to be assessed. While the removal of attached epifauna can lead to changes in species composition or reductions in the productivity of fish populations, the magnitude and direction of such effects critically depend on substrate type, depth, fishing intensity, and other factors (see below).
Short-term Impacts on Benthic Communities
Potential short-term impacts of trawling on benthic communities (infauna and epifauna) include:
- physical damage to benthic organisms;
- direct mortality of vulnerable species;
- reduction in diversity and abundance of some taxa;
- increase of scavengers and other opportunistic species in disturbed areas.
Studies of such short-term impacts have yielded variable results as summarized in Auster and Langton (1999). Benthic communities in shallow, soft-bottom, high-energy areas are typically resilient because natural levels of disturbance are high, but short-term changes in community composition or in the abundance of individual groups have been documented in many cases. A recent study examined the impact of a commercial otter trawl on the benthic infauna in muddy sediment in the Northwest Mediterranean Sea at 30-40 m immediately before and at several intervals (1 – 150 hours) after trawling (Sánchez et al. 2000). The only observed effect was an apparent decrease in the total number of individuals at the unfished control site 150 hours after fishing, a result that cannot be attributed to trawling. The authors also report differences over time in relative community composition, but these differences were common to both the fished and unfished sites (Sánchez et al. 2000). Other studies have shown increases in opportunistic species (primarily polychaetes), and decreases in bivalves and other species with limited mobility (Tuck et al. 1998, Ball et al. 2000, Bergman and Santbrink 2000b). However, many experiments have revealed large temporal changes in abundance and/or diversity measures at both trawled and reference sites that often exceeded the variations attributed to trawling (e.g. Tuck et al. 1998). Furthermore, differences in community structure are often apparent prior to treatment and benthic community composition changes over time at both trawled and reference sites (e.g. Tuck et al. 1998). These results are broadly consistent with previous reviews (Jennings 1998, Auster and Langton 1999), but have added an understanding of the importance of temporal changes in benthic communities that occur independent of fishing.
Trawl impacts are generally believed to be larger in deeper, more stable habitats (Auster and Langton 1999), although few experiments have been conducted below 100 m. One exception is a recent 3-year study that may be particularly relevant to Alaska because it used a commercial trawl similar to those used in Alaska and sampled at depths of 120-146 m, which is similar to the depth range (101-200 m) that experienced the highest bottom trawl duration in the Gulf of Alaska (PSEIS, section 184.108.40.206). The study used a careful experimental design to examine the effects of high-density trawling on the benthic community. The substrate in the study area consisted of fine to medium-grained sand and was relatively stable, although interannual variations in sediment characteristics were apparent, possibly due to winter storms (Schwinghamer et al. 1998). Immediate effects of trawling included a reduction in the biomass of invertebrate bycatch over 12 sets (Prena et al. 1999). The biomass of epibenthic organisms was generally lower in trawled corridors compared to nearby reference sites, which was attributed to direct removals by the trawl, mortality or damage and subsequent predation, as well as migration. However, differences between trawled and reference sites were apparent prior to trawling and large trends over time were apparent at both trawled and reference sites. Kenchington et al. (2001) examined the effects of the same trawling experiment on benthic macrofauna. Immediate effects of trawling were only observed in 1994, when the abundance and/or biomass of 15 species (mostly polychaetes) were significantly lower after trawling. No effects were found in 1993 or 1995. Similarly, no immediate effects of trawling on overall community composition were found in either 1993 or 1995. These results suggest that trawling impacts in deeper areas are not necessarily more severe that in shallow areas.
Many of the above studies point to an important caveat for interpreting trawl impact studies that was not discussed by Auster and Langton (1999). Studies comparing trawled and untrawled areas, as well as before-after studies, are prone to misinterpretation because benthic communities are characterized by large spatial and temporal variations that are often difficult to distinguish from trawling impacts (Hall 1999). For example, several studies (Tuck et al. 1998, Prena et al. 1999, Kenchington et al. 2001) revealed natural variations in benthic community composition on relatively short temporal scales (< 1 year) that often exceeded the magnitude of trawling effects. Kenchington et al. (2001) concluded that observed trawling disturbances appeared to mimic natural disturbance at reference sites. Therefore, observed changes in macrobenthic community composition could not be attributed to trawling. Similarly, Lindegarth et al. (2000a) observed large changes before and after trawling at both an untrawled site and a site trawled repeatedly using a commercial shrimp trawl over a period of 12 months. Natural variability at both sites greatly exceeded any effects of trawling. The same caveat applies to studies that compare heavily fished and unfished sites, which may be difficult to separate from natural gradients on the sea floor (Hall 1999). Lindegarth et al. (2000b) provided empirical evidence that spatial confounding may cause serious problems in the interpretation of experiments with only one control and one trawled area. Many comparative studies fail to take into account pre-existing differences between trawled and untrawled sites and may erroneously attribute differences to trawling effects. Clearly, results from trawl impact studies have to be interpreted in light of natural variability and the magnitude of effects in many cases may be well within the range of natural variability.
While short-term effects have been documented in a number of cases, the ecological significance of such changes is largely unknown. As Eleftheriou (2000) pointed out, merely showing an impact does not necessarily imply ecosystem damage. In recent years, research has moved from simply showing an impact to directly estimating trawling mortality for various groups of benthic organisms due to the passage of a trawl. Bergman and Santbrink (2000b) estimated mortalities for various benthic organisms with limited mobility and found that mortalities for two types of beam trawl ranged from 5-40% for most species and 20-65% for some bivalves, but were substantially lower for an otter trawl. Bergman and Santbrink (2000a) used these estimates together with effort data and maps of the distribution of benthic organisms to compute annual mortalities for various taxa in the Dutch sector of the North Sea. Annual mortalities due to all trawling ranged from 5-39% and were mostly due to the 12 m beam trawl, which is the most frequently used gear in this area and inflicted the highest mortality rates of the gears tested. This and similar studies (Moran and Stephenson 2000, Piet et al. 2000, Pitcher et al. 2000) are a first step towards quantifying impacts of trawling on non-target species, including fishes and invertebrates, and towards effectively managing non-target populations. However, very little is known about the population biology of most of the affected species and it is currently unclear what levels of fishing-induced mortality are sustainable. Alternatively, long-term studies of species removals in relation to fishing effort can help to estimate levels of sustainable fishing intensity (Pitcher et al. 2000, see below).
A major concern has been the reduction in species diversity that has been observed in some short-term fishing experiments (Auster and Langton 1999). Recent studies are consistent with these results and have shown reductions in diversity measures of the benthic community. For example, otter trawling at shallow, muddy sites (Tuck et al. 1998, Ball et al. 2000), as well as a deep, muddy site (Smith et al. 2000) resulted in short-term reduction in species diversity. However, other studies failed to detect effects on diversity (Prena et al. 1999, Kenchington et al. 2001). The interpretation of reductions in diversity is not always clear and may be the result of increases in less abundant species or decreases in abundant species (Hall 1999). Often, decreases in diversity result from the disproportionate increase in the abundance of opportunistic species due to immigration. Moreover, the role of diversity in maintaining the productivity or stability of marine communities is largely unknown (Jennings and Reynolds 2000). These authors concluded that diversity measures are not particularly useful or sensitive indicators; instead they suggest identifying indicator species that are vulnerable to fishing.
Recovery rates of benthic invertebrates are poorly understood and depend on the population biology of the affected species. Recovery rates will also vary among substrate and natural levels of disturbance. A meta-analysis of trawl impact studies (Collie et al. 2000) suggests that recovery rates on sand are on the order of 100 days, suggesting that sandy areas could tolerate approximately 3 trawling events per year on average. Although data for other soft-bottom sediments were sparse, similar recovery rates were indicated for mud and muddy sand (Collie et al. 2000). Clearly, recovery is also dependent on the spatial distribution of fishing effort because populations of mobile species recover in part due to immigration from unfished areas (Jennings et al. 2001). This again emphasizes the importance of understanding the small-scale distribution of fishing effort.
With the exception of two otter trawl studies, Auster and Langton (1999) examined short-term effects only for relatively heavy gear such as dredges and beam trawls. Therefore, conclusions from many of the cited studies cannot be generalized to otter trawl gear such as the commercial gear used in Alaska. Several studies indicate that otter trawls tend to have less impact on benthos than beam trawls and dredges (Bergman and Santbrink 2000b, Collie et al. 2000, see below).
In summary, short-term effects of trawling on benthic communities have been demonstrated in various habitats and general conclusions in Auster and Langton (1999) on the potential short-term impacts have been confirmed by experimental studies. However, some earlier studies failed to properly account for natural variability among sites and over time. It has recently become clear that benthic communities are highly dynamic and can undergo large changes regardless of fishing activity. In general, the response of benthic organisms to disturbance differs by gear, substrate, area, and type of organism (Collie et al. 2000). Shallow areas characterized by sand and relatively small and motile organisms are often highly resilient and recover quickly from trawling. Where impacts have been demonstrated, infaunal and epifaunal bivalves, gastropods, and other organisms with limited mobility typically experience some degree of direct mortality or damage from trawl gear. These become prey for mobile benthic predators and scavengers, whose abundances often increase in response to trawling but quickly return to pre-trawling levels. Simply showing impacts of trawl gear on habitat is not sufficient for evaluating the wider ecological consequences of trawling. Recent studies that estimate fishing-induced mortality for vulnerable species may help to determine sustainable levels of fishing mortality for benthic organisms. However, such studies are currently not feasible for the Gulf of Alaska or Bering Sea because data on the distribution of most benthic organisms, as well as data on the small-scale distribution of fishing effort is not available.
Long-term Impacts on Benthic Communities
Arguably the most important types of fishing gear impacts are long-term changes in community composition and other effects on the larger ecosystem, including effects on non-target and harvested fish species. These are also the least understood and most controversial aspects of fishing gear effects. Many of the observed long-term changes in benthic habitats and communities may not be a result of fishing activity per se, but of the removal of target (as well as bycatch) species. Such effects are often difficult to distinguish from effects of habitat modifications.
Potential long-term impacts of trawling and fixed gear on benthic communities include:
- long-term trends in the abundance of some benthic species (both increases and decreases in abundance);
- long-term trends in species diversity of the benthic community;
- long-term trends in the productivity of an ecosystem;
- long-term changes in benthic community composition.
Most of the “long-term” studies reviewed by Auster and Langton (1999) assess changes over several months or at most a few years. While increases and decreases in different components of the benthic community were reported for many studies, several qualifications have to be added in interpreting these studies.
As pointed out previously, conclusions regarding the relative importance of fishing effects in explaining observed changes are often equivocal, particularly in the light of natural variations in the abundance of many species (Hall 1999, Lindegarth et al. 2000b). For example, Arntz et al. (1994), as cited in Auster and Langton (1999, Table 5), compared communities around a sunken ship (presumably unfished) with an adjacent, heavily fished area. While differences in abundance between sites and trends in some species in the fished area were found, they cannot be attributed to fishing based on a comparison of only two sites. Auster and Langton (1999, Table 5) further cite Currie and Parry (1996) as showing long-term impacts on some species after dredging. However, multivariate analysis in this study indicated that changes to community structure caused by dredging were generally smaller than differences between seasons and years (Currie and Parry 1996). Prena et al. (1996) are cited as documenting a decrease in invertebrate bycatch after 3 years of trawling. However, a strong decrease in benthic biomass was observed at both trawled and reference sites (Prena et al. 1999, Kenchington et al. 2001), suggesting that fishing was not responsible for the observed long-term changes. Collie et al. (1997) documented a gradient of community structure from deep undisturbed to shallow disturbed sites (Auster and Langton 1999, Table 5). Clearly, depth and fishing disturbance were highly confounded; therefore effects cannot unequivocally be attributed to dredging. Given that benthic community composition generally differs most strongly along the depth gradient (Mueter and Norcross 1999), large differences with depth can be expected even in the absence of fishing. Furthermore, disturbed sites were dominated by mollusks, crabs, and echinoderms (Collie et al. 1997), which is not consistent with other studies that suggest that mollusks and crabs should decrease in response to fishing (e.g. Bergman and Santbrink 2000a). In general, studies that document differences in benthic community composition along a fishing gradient at one point in time are often inconclusive. The reason that fishing is concentrated in certain areas is likely due to the fact that these areas have higher abundances of desired species and other unique characteristics to begin with (Hall 1999).
Only 6 of 25 long-term studies reviewed in Auster and Langton (1999) examine changes over more than 5 years and can be considered truly long-term. All of these studies report increases and/or decreases in some components of the benthic community, but it is not clear whether the observed changes are the result of fishing. Three of these studies (Reise 1982, Riesen and Reise 1982, Holme 1983) were reviewed in Hall (1999), who found that the observed changes reported in all of these studies could not be attributed to fishing effects. Hall (1999) also reviewed other evidence for long-term trends in benthic communities caused by fishing in the North Sea and concluded that “the case for invoking fisheries as a primary cause for the recorded changes is not very strong” (his emphasis). In particular, fisheries impacts are too often confounded with natural effects. In other cases, observed long-term changes are opposite to what would be expected if fishing were the cause, for example an increase in “vulnerable” species over time (Kröncke 1990, Frid et al. 2000).
The above examples clearly illustrate the difficulties of identifying long-term impacts of fishing gear on benthic communities and suggest that in at least some of the studies reviewed by Auster and Langton (1999) fishing effects were not responsible for the observed changes. A number of recent studies have examined long-term impacts (> 10 years) of trawling in the North Sea (Craeymeersch et al. 2000, Frid and Clark 2000, Frid et al. 2000, Kaiser et al. 2000, Ramsay et al. 2000, Rumohr and Kujawski 2000), off New Zealand (Thrush et al. 1998), off Australia (Pitcher et al. 2000), and in the Bering Sea (McConnaughey et al. 2000).
I first review recent North Sea studies. Frid et al. (2000) found that the abundance of a number of taxa (including both opportunistic species and species vulnerable to fishing) increased on 3 out of 5 fishing grounds between 1923-25 and the 1980s. However, different gears were used in the two periods, which may explain some of the observed differences. Furthermore, an increase in sensitive species is not consistent with trawl impacts, suggesting that the changes are just as likely to be due to natural variability. Nevertheless, Frid and Clark (2000), after reviewing the evidence, concluded that long-term changes have taken place in North Sea benthic communities, notably decreases in the abundance of bivalves and increases in the abundance of scavenging crustaceans and sea stars. Craeymeersch et al. (2000) examined benthic community composition in relation to the fine-scale distribution of fishing effort and found significant differences in species composition between heavily fished and less heavily fished areas. However, trawling effort explained only 0.9 to 2.2% of the variability in species composition and the major part of differences in species composition was related to environmental factors. Abundances of several opportunistic species showed no consistent differences between heavily fished and lightly fished areas (Craeymeersch et al. 2000).
Two studies on the Australian shelf (Pitcher et al. 2000) and in the Hauraki Gulf, New Zealand (Thrush et al. 1998) have shown differences in community composition related to otter trawling intensity. For example, on the Australian shelf, the community composition in heavily fished areas shifted to a less vulnerable community (Pitcher et al. 2000). Nevertheless, the authors conclude that impacts of trawling near the Great Barrier Reef have the potential to be ecologically sustainable if they are appropriately managed. Thrush et al. (1998) also tested the a priori predictions that scavengers would be more abundant at heavily fished sites. However, no effect of fishing pressure on scavengers was found, suggesting that such effects are transient. Ramsay et al. (2000) similarly examined the a priori hypothesis that the invertebrate scavenger Asterias rubens would be more abundant in areas of the southern North Sea that have a higher fishing effort. They found a weak quadratic relationship between A. rubens abundance and fishing effort, suggesting that starfish numbers first increased, then decreased with further increases in fishing pressure. However, the model explained only 2.5% and 8% of the variability in the two study regions, suggesting that other factors affect starfish abundances much more strongly.
It is generally believed that fishing has reduced benthic diversity in some areas due to direct fishing mortality, loss of sessile epifauna such as bryozoans and hydroids, and decreases in the abundance of some rare species (Kaiser and Spencer 1996, Collie et al. 1997, Jennings and Reynolds 2000). Changes in diversity are likely to be more pronounced on hard substrates with attached fauna (Collie et al. 1997), whereas there is no evidence for effects on biogenic structures and associated fauna that are not attached to the substrate (Kaiser et al. 1999). To evaluate the impacts of trawling on overall diversity of a system we first need to understand the fine-scale distribution of fishing effort.
While most studies of trawling impacts focus on benthic invertebrates, an equally or more important concern is whether trawling may cause changes in the associated fish fauna. While the species composition of demersal fish communities on many continental shelf areas (e.g. North Sea, Georges Bank, Gulf of Thailand) has undergone large changes over time (for case studies, see Hall 1999), such changes have not been linked to the effects of fishing gear on habitat. Trends in species composition of fishes can rarely be attributed to a single cause (e.g. Greenstreet and Rogers 2000). They are more likely to result from a combination of the direct removal of target and non-target species and environmental changes than from indirect effects of habitat modifications (see case studies in Hall 1999).
Similarly, long-term effects of fixed gear on benthic communities are likely to occur through the removal of target and bycatch species and associated changes in species composition and trophic interactions. Therefore, the effects will be similar to those of mobile fishing gear and depend largely on the rates of bycatch and discarding, which can be substantial for both types of gear (Alverson et al. 1994). Additional concerns arise from the loss of fixed gear (pots or longlines) and their potential for ghost fishing, which can have large impacts on both target and non-target populations (e.g. Kruse and Kimber 1993, see below).
In summary, the evidence for long-term changes due to the impacts of fishing gear on benthic communities is relatively weak and often inconclusive. The strongest evidence for effects on continental shelf systems is from areas in the Southern hemisphere with coral reefs that are characterized by large attached epifauna (e.g. Pitcher et al. 2000). Intense fishing in such areas has the potential for causing a shift to communities dominated by smaller, more resilient species (e.g. Kaiser et al. 2000). One of the most important factors determining the magnitude and direction of long-term changes is likely to be fishing intensity (e.g. Poiner et al. 2000, cited in Pitcher et al. 2000). The challenge for management is to determine levels of fishing that are sustainable and will not degrade benthic habitat in the long run.
Can Results be Generalized?
While fishing impact studies have allowed some general conclusions about the effects of fishing gear on habitat and benthic communities, it is critically important to consider substrate type, depth, gear type, type of organism, and fishing intensity when evaluating impacts on benthos. For example, the magnitude of effects of any given gear differs among substrates, while for any given substrate, effects of different gears differ widely (Collie et al. 2000). I will briefly discuss the role of these factors in determining the effects of trawling. While the factors are discussed separately, they are often confounded (for example, depth and sediment type).
- Substrate type: There is a generally accepted gradient of vulnerability from coarse sand to fine sand to mud to gravel and mixed grounds (Collie et al. 2000, Moore and Jennings 2000). Substrates such as muddy sand, sandy mud, and sand typically found in the Bering Sea and in parts of the Gulf of Alaska are least vulnerable and have the shortest recovery times. Hard-bottom substrates with attached epifauna are clearly vulnerable to both mobile and fixed fishing gear (although the magnitude of the effects for fixed gear is unknown). To quantify potential impacts, information on substrate distribution is critically important, but is currently lacking for much of the Gulf of Alaska.
- Depth: Generally, impacts tend to be smaller in shallow, high, energy environments where the level of natural disturbance is high. Few studies have been done in deep areas, but Schwinghamer et al. (1998) and Kenchington et al. (2001) found no longer-term effects of high intensity otter trawling on sediment characteristics and infaunal organisms at a relatively deep (120 – 146 m) site, suggesting that deep areas are not necessarily more vulnerable. Very deep, stable environments are probably most vulnerable, but to my knowledge no studies have been conducted below 200 m to date.
- Gear type: Mobile fishing gears are generally considered to have more pronounced effects on habitat and benthic communities than fixed gears and have therefore been studied most extensively. With two exceptions, the studies of short-term impacts reviewed by Auster and Langton (1999) used beam trawls and scallop or clam dredges, thus conclusions from these studies are not generally applicable to Alaska. Dredges can have large impacts (e.g. Thrush et al. 1998), yet recovery times may be fast in shallow areas subject to intense wave and storm activity. Comparisons between beam trawls and otter trawls suggest that the impact of otter trawls is generally less. Bergman and Santbrink (2000b) found that mortalities of infaunal and epifaunal organisms were considerably lower for an otter trawl than for a beam trawl on sandy substrate. Similarly, among all mobile gears (dredges, beam trawls, otter trawls), otter trawls such as those used in the Bering Sea and Gulf of Alaska have been found to be the least damaging (Collie et al. 2000). In addition, there is some potential for reducing the effects of otter trawls further by making appropriate modifications (Moran and Stephenson 2000).
- Type of organism: Infauna and epifauna in shallow, high-energy areas appear highly resilient, while large sessile epifauna is most vulnerable (Auster and Langton 1999, ICES 2000, Kaiser and de Groot 2000). Other sessile organisms or those with limited mobility such as bivalves and some crustaceans can experience relatively high mortalities from trawling (see Bergman and Santbrink 2000b), but dead and dying animals are rapidly consumed by scavengers. Motile epifauna and small infauna is least vulnerable.
- Fishing intensity: An assessment of fishing intensity on fine spatial scales is critically important in evaluating the overall impact of fishing gear on different habitats and may be achieved, for example, by satellite tracking of fishing vessels (Jennings et al. 2000). Studies of fine-scale effort distribution have shown the highly localized nature of trawling operations (Rijnsdorp et al. 1998). The more localized fishing effort is, the larger is the area that is rarely or never fished. It has been shown that low-intensity fishing may have minor impacts, while cumulative impacts of high-intensity trawling can be substantial (Poiner et al. 2000, cited in Pitcher et al. 2000). The small-scale distribution of fishing effort is currently not known for either the Bering Sea or the Gulf of Alaska. However, the average number of trawls per unit area in the Gulf of Alaska and Aleutian Islands (PSEIS, Fig 3.2-3, 3.2-4) suggests a relatively low trawl density in this area compared to other shelf areas around the world (Rijnsdorp et al. 1998, Piet et al. 2000, Pitcher et al. 2000).
While many of the general findings of Auster and Langton (1999) on short-term effects of fishing on habitat and benthic communities have been confirmed by more recent studies, there is little evidence for long-term changes as a result of fishing in most habitats. Many of the long-term studies cited in Auster and Langton (1999) were flawed or can be interpreted in different ways. Given that long-term impacts depend strongly on gear type and intensity of fishing, the potential for significant long-term impacts of trawl fishing on benthic communities appears to be lower in the Gulf of Alaska and Bering Sea compared to many other areas.
Similarly, many of the conclusions of Auster and Langton (1999) cannot be generalized to Alaska. Most of the short-term studies reviewed were on hard-bottom or vulnerable biogenic sediment, whereas most of the trawl fishing grounds in Alaska, particularly in the Bering Sea, are on less vulnerable substrate. However, the distribution of different substrates in the Gulf of Alaska is largely unknown. As well, the otter trawls used in the groundfish fishery off Alaska are the least damaging gear type among the mobile bottom fishing gears examined to date. Finally, fishing intensities in the Gulf of Alaska and Bering Sea are likely to be less than in other heavily fished areas such as the North Sea and the Western Atlantic.
Furthermore, any changes in benthic communities due to fishing must be evaluated in the context of natural variability. Recent studies clearly indicate that benthic communities are highly dynamic and that natural variability often exceeds any effects of fishing. This includes large changes over time as well as large spatial variability due to underlying environmental gradients.
While many studies have shown statistically significant effects, this does not imply ecologically important effects. For example, it is not clear whether the observed differences between unfished and heavily fished areas in the Bering Sea (McConnaughey et al. 2000) are ecologically significant. Ultimately, it is more important to estimate effect sizes and use these to determine levels of fishing intensity for a given habitat that are sustainable. This requires estimates of the fine-scale distribution of fishing effort and knowledge of the distribution of benthic organisms. Alternatively, analyses of benthic community structure in relation to fishing effort combined with modeling can provide estimates of sustainable fishing rates (Pitcher et al. 2000). The analysis in the PSEIS on bycatch rates of different gears is a step in the right direction. However, such models need to take into account the distribution of effort at a finer scale and any changes in effort distribution resulting from different allocations of catches among gears.
A review of the paper by Auster and Langton (1999) and of more recent literature on the impacts of trawling suggests that the paper is not an adequate summary of trawling impacts for evaluating potential benefits from management measures in Alternative 5 of the PSEIS. Auster and Langton (1999) reviewed a selected set of short-term studies that are not representative of the Bering Sea and Gulf of Alaska trawl fisheries, neither in terms of gear types reviewed, nor, most likely, in terms of the substrate on which trawling typically occurs. Therefore, the conclusion that a reduction of trawling in Alaskan waters and increased allocation of catches to fixed gear will have definite benefits in terms of increased habitat complexity and species diversity is not tenable. In addition, the shortcomings of some of the long-term studies cited in Auster and Langton (1999), suggest that there is little evidence for long-term effects even in some heavily fished areas. Therefore anticipated benefits from the management measures in Alternative 5 may not be as expected.
- Alverson, D. L., M. H. Freeberg, S. A. Murawski, and J. G. Pope. 1994. A global assessment of bycatch and discards. FAO Fisheries Technical Paper, Rome.
- Auster, P. J., and R. W. Langton. 1999. The effects of fishing on fish habitat. American Fisheries Society Symposium 22:150-187.
- Ball, B. J., G. Fox, and B. W. Munday. 2000. Long- and short-term consequences of a Nephrops trawl fishery on the benthos and environment of the Irish Sea. ICES Journal of Marine Science 57:1315-1320.
- Bergman, M. J. N., and J. W. V. Santbrink. 2000a. Fishing mortality of populations of megafauna in sandy sediments. Pages 49-68 in M. J. Kaiser and S. J. de Groot, editors. Effects of fishing on non-target species and habitats: biological, conservation, and socio-economic issues. Blackwell Science, Oxford.
- Bergman, M. J. N., and J. W. V. Santbrink. 2000b. Mortality in megafaunal benthic populations caused by trawl fisheries on the Dutch continental shelf in the North Sea in 1994. ICES Journal of Marine Science 57:1321-1331.
- Brylinski, M., J. Gibson, and D. C. Gordon Jr. 1994. Impacts of flounder trawls on the intertidal habitat and community of the Minas Basin, Bay of Fundy. Canadian Journal of Fisheries and Aquatic Sciences 51:650-661.
- Churchill, J. H. 1989. The effect of commerical trawling on sediment resuspension and transport over the Middle Atlantic Bight continental shelf. Continental Shelf Research 9:841-64.
- Collie, J. S., G. A. Escanero, and P. C. Valentine. 1997. Effects of bottom fishing on the benthic megafauna of Georges Bank. Marine Ecology Progress Series 155:159-172.
- Collie, J. S., S. J. Hall, M. J. Kaiser, and I. R. Poiner. 2000. A quantitative analysis of fishing impacts on shelf-sea benthos. Journal of Animal Ecology 69:785-798.
- Craeymeersch, J. A., G. J. Piet, A. D. Rijnsdorp, and J. Buijs. 2000. Distribution of macrofauna in relation to the micro-distribution of trawling effort. Pages 187-197 in M. J. Kaiser and S. J. de Groot, editors. Effects of fishing on non-target species and habitats: biological, conservation, and socio-economic issues. Blackwell Science, Oxford.
- Currie, D. R., and G. D. Parry. 1996. Effects of scallop dredging on a soft sediment community: a large-scale experimental study. Marine Ecology Progress Series 134:131-150.
- DeAlteris, J., L. Skrobe, and C. Lipsky. 1999. The significance of seabed disturbance by mobile fishing gear relative to natural processes: a case study in Narragansett Bay, Rhode Island. American Fisheries Society Symposium 22:224-237.
- Eleftheriou, A. 2000. Marine benthos dynamics: environmental and fisheries impacts: Introduction and Overview. ICES Journal of Marine Science 57:1299-1302.
- Fonteyne, R. 2000. Physical impact of beam trawls on seabed sediments. Pages 15-36 in M. J. Kaiser and S. J. de Groot, editors. Effects of fishing on non-target species and habitats: biological, conservation, and socio-economic issues. Blackwell Science, Oxford.
- Freese, L., P. J. Auster, J. Heifetz, and B. L. Wing. 1999. Effects of trawling on seafloor habitat and associated invertebrate taxa in the Gulf of Alaska. Marine Ecology Progress Series 182:119-126.
- Frid, C. L. J., and R. A. Clark. 2000. Long-term changes in North Sea benthos: discerning the role of fisheries. Pages 198-216 in M. J. Kaiser and S. J. de Groot, editors. Effects of fishing on non-target species and habitats: biological, conservation, and socio-economic issues. Blackwell Science, Oxford.
- Frid, C. L. J., K. G. Harwood, S. J. Hall, and J. A. Hall. 2000. Long-term changes in the benthic communities on North Sea fishing grounds. ICES Journal of Marine Science 57:1303-1309.
- Greenstreet, S. P. R., and S. I. Rogers. 2000. Effects of fishing on non-target species. Pages 217-236 in M. J. Kaiser and S. J. de Groot, editors. Effects of fishing on non-target species and habitats: biological, conservation, and socio-economic issues. Blackwell Science, Oxford.
- Hall, S. J. 1999. The effects of fishing on marine ecosystems and communities. Blackwell Science, Oxford, UK.
- Hampton, M. A., P. R. Carlson, H. J. Lee, and R. A. Feely. 1986. Geomorphology, sediment, and sedimentary processes. Pages 93-143 in D. W. Hood and S. T. Zimmerman, editors. The Gulf of Alaska: Physical environment and biological resources. Minerals Management Service, Springfield, VA.
- Holme, N. A. 1983. Fluctuations in the benthos of the western English Channel. Oceanologica Acta XX:121-124.
- Jennings, S. 1998. The effects of fishing on marine ecosystems. Pages 201-352 in J. H. S. Blaxter, A. J. Southward, and P. A. Tyler, editors. Advances in Marine Biology. Academic Press, San Diego.
- Jennings, S., J. K. Pinnegar, N. V. C. Polunin, and K. J. Warr. 2001. Impacts of trawling disturbance on the trophic structure of benthic invertebrate communities. Marine Ecology Progress Series 213:127-142.
- Jennings, S., and J. D. Reynolds. 2000. Impacts of fishing on diversity: from pattern to process. Pages 235-250 in M. J. Kaiser and S. J. de Groot, editors. Effects of fishing on non-target species and habitats: biological, conservation, and socio-economic issues. Blackwell Science, Oxford.
- Jennings, S., K. J. Warr, S. P. R. Greenstreet, and A. J. R. Cotter. 2000. Spatial and temporal patterns in North Sea fishing effort. in M. J. Kaiser and S. J. de Groot, editors. Effects of fishing on non-target species and habitats: biological, conservation, and socio-economic issues. Blackwell Science, Oxford.
- Kaiser, M. J., and S. J. de Groot, editors. 2000. Effects of fishing on non-target species and habitats: biological, conservation, and socio-economic issues. Blackwell Science, Oxford.
- Kaiser, M. J., D. B. Edwards, P. J. Armstrong, K. Radford, N. E. L. Lough, R. P. Flatt, and H. D. Jones. 1998. Changes in megafaunal benthic communities in different habitats after trawling disturbance. ICES Journal of Marine Science 55:353-361.
- Kaiser, M. J., K. Ramsay, C. A. Richardson, F. E. Spence, and A. R. Brand. 2000. Chronic historical fishing disturbance has changed shelf sea benthic community structure. Journal of Animal Ecology 69:494-503.
- Kaiser, M. J., S. I. Rogers, and J. R. Ellis. 1999. Importance of benthic habitat complexity for demersal fish assemblages. American Fisheries Society Symposium 22:212-223.
- Kaiser, M. J., and B. E. Spencer. 1996. The effects of beam-trawl disturbance on infaunal comunities in different habitats. Journal of Animal Ecology 65:348-358.
- Kenchington, E. L. R., J. Prena, K. D. Gilkinson, D. C. G. Jr., K. MacIsaac, C. Bourbonnais, P. J. Schwinghamer, T. W. Rowell, D. L. McKeown, and W. P. Vass. 2001. Effects of experimental otter trawling on the macrofauna of a sandy bottom ecosystem on the Grand Banks of Newfoundland. Canadian Journal of Fisheries and Aquatic Sciences 58:1043-1057.
- Kröncke, I. 1990. Macrofauna standing stock of the Dogger Bank. A comparison: II. 1951-1952 versus 1985-1987. Are changes in the community of the northeastern part of the Dogger Bank due to environmental changes? Netherlands Journal of Sea Research 25:189-198.
- Kruse, G. H., and A. Kimber. 1993. Degradable escape mechanisms for pot gear: a summary report to the Alaska Board of Fisheries. Alaska Department of Fish and Game, Juneau.
- Lindegarth, M., D. Valentinsson, M. Hansson, and M. Ulmestrand. 2000a. Effects of trawling disturbances on temporal and spatial structure of benthic soft-sediment assemblages in Gullmarsfjorden, Sweden. ICES Journal of Marine Science 57:1369-1376.
- Lindegarth, M., D. Valentinsson, M. Hansson, and M. Ulmestrand. 2000b. Interpreting large-scale experiments of effects of trawling on benthic fauna: an empirical test of the potential effects of spatial confounding in experiments without replicated control and trawled areas. Journal of Experimental Marine Biology and Ecology 245:155-169.
- Marlow, M. S., A. J. Stevenson, H. Chezar, and R. A. McConnaughey. 1999. Tidally generated sea-floor lineations in Bristol Bay, Alaska. Geo-Marine Letters 19:219-226.
- McConnaughey, R. A., K. L. Mier, and C. B. Dew. 2000. An examination of chronic trawling effects on soft-bottom benthos of the eastern Bering Sea. ICES Journal of Marine Science 57:1377-1388.
- Moore, G., and S. Jennings, editors. 2000. Commercial fishing: the wider ecological impacts. British Ecological Society, Blackwell Science, Cambridge.
- Moran, M. J., and P. C. Stephenson. 2000. Effects of otter trawling on macrobenthos and management of demersal scalefish fisheries on the continental shelf of north-western Australia. ICES Journal of Marine Science 57:510-516.
- Mueter, F. J. 1999. Spatial and temporal patterns in the Gulf of Alaska groundfish community in relation to the environment. Ph.D. University of Alaska Fairbanks, Fairbanks.
- Mueter, F. J., and B. L. Norcross. 1999. Linking community structure of small demersal fishes around Kodiak Island, Alaska, to environmental variables. Marine Ecology Progress Series 190:37-51.
- Piet, G. J., A. D. Rijnsdorp, M. J. N. Bergman, J. W. v. Santbrink, J. Craeymeersch, and J. Buijs. 2000. A quantitative evaluation of the impact of beam trawling on benthic fauna in the southern North Sea. ICES Journal of Marine Science 57:1332-1339.
- Pitcher, C. R., I. R. Poiner, B. J. Hill, and C. Y. Burridge. 2000. Implications of the effects of trawling on sessile megazoobenthos on a tropical shelf in northeastern Australia. ICES Journal of Marine Science 57:1359-1368.
- Prena, J., P. Schwinghamer, T. W. Rowell, D. C. Gordon Jr, K. D. Gilkinson, W. P. Vass, and D. L. McKeown. 1999. Experimental otter trawling on a sandy bottom ecosystem of the Grand Banks of Newfoundland: analysis of trawl bycatch and effects on epifauna. Marine Ecology Progress Series 181:107-124.
- Ramsay, K., M. J. Kaiser, A. D. Rijnsdorp, J. A. Craeymeersch, and J. Ellis. 2000. Impact of trawling on populations of the invertebrate scavenger Asterias rubens. Pages 151-162 in M. J. Kaiser and S. J. de Groot, editors. Effects of fishing on non-target species and habitats: biological, conservation, and socio-economic issues. Blackwell Science, Oxford.
- Reise, K. 1982. Long-term changes in the macrobenthic invertebrate fauna of the Wadden Sea: are polychaetes about to take over? Netherlands Journal of Sea Research 16:29-36.
- Riesen, W., and K. Reise. 1982. Macrobenthos of the subtidal Wadden Sea: revisited after 55 years. Helgoländer Meeresuntersuchungen 35:409-423.
- Rijnsdorp, A. D., A. M. Buys, F. Storbeck, and E. G. Visser. 1998. Micro-scale distribution of beam trawl effort in the southern North Sea between 1993 and 1996 in relation to the trawling frequency of the sea bed and the impact on benthic organisms. ICES Journal of Marine Science 55:403-419.
- Rumohr, H., and T. Kujawski. 2000. The impact of trawl fishery on the epifauna of the southern North Sea. ICES Journal of Marine Science 57:1389-1394.
- Sainsbury, K. J., R. A. Campbell, R. Lindholm, and A. W. Whitelaw. 1997. Experimental management of a multispecies fishery: examining the possibility of trawl-induced habitat modification in E. K. Pikitch, D. D. Huppert, and M. P. Sissenwine, editors. Global trends: fisheries management. American Fisheries Society, Bethesda, Maryland.
- Sánchez, P., M. Demestre, M. Ramon, and M. J. Kaiser. 2000. The impact of otter trawling on mud communities in the northwestern Mediterranean. ICES Journal of Marine Science 57:1352-1358.
- Schwinghamer, P., D. C. Gordon Jr., T. W. Rowell, J. Prena, D. L. McKeown, G. Sonnichsen, and J. Y. Guigné. 1998. Effects of experimental otter trawling on surficial sediment properties of a sandy-bottom ecosystem on the Grand Banks of Newfoundland. Conservation Biology 12:1215-1222.
- Smith, C. J., K. N. Papadopoulou, and S. Diliberto. 2000. Impact of otter trawling on an eastern Mediterranean commercial trawl fishing ground. ICES Journal of Marine Science 57:1340-1351.
- Thrush, S. F., J. E. Hewitt, V. J. Cummings, P. K. Dayton, M. Cryer, S. J. Turner, G. A. Funnell, R. G. Budd, C. J. Milburn, and M. R. Wilkinson. 1998. Disturbance of the Marine Benthic Habitat by Commercial Fishing: Impacts at the Scale of the Fishery. Ecological Applications 8:866-879.
- Tuck, I. D., S. J. Hall, M. R. Robertson, E. Armstrong, and D. J. Basford. 1998. Effects of physical trawling disturbance in a previously unfished sheltered Scottish sea loch. Marine Ecology Progress Series 162:227-242.
- Van Dolah, R. F., P. H. Wendt, and N. Nicholson. 1987. Effects of a research trawl on a hard- bottom assemblage of sponges and corals. Fisheries Research 5:39-54.