ICES Journal of Marine Science: Journal du Conseil Advance Access originally published online on October 25, 2006
ICES Journal of Marine Science: Journal du Conseil 2007 64(1):110-121; doi:10.1093/icesjms/fsl006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Potential pressure indicators for fishing, and their data requirements
1 Wageningen IMARES, PO Box 68, 1970 AB IJmuiden, The Netherlands
2 Fisheries Research Services, Marine Laboratory, PO Box 101, Victoria Road, Aberdeen, AB11 9DB, UK
Correspondence to G. J. Piet: tel: +31 255 564699; fax: +31 255 564644; e-mail: gerjan.piet{at}wur.nl
Piet, G. J., Quirijns, F. J., Robinson, L., and Greenstreet, S. P. R. 2007. Potential pressure indicators for fishing, and their data requirements. – ICES Journal of Marine Science, 64: 110–121.Indicators of fishing pressure are necessary to support an ecosystem approach to fisheries management (EAFM). We present a framework that distinguishes four levels of pressure indicators that move from being a simple description of anthropogenic activity to more precisely describing the actual pressure on the ecosystem and its components, but which require increasingly more information to be quantified. We use the example of the Dutch beam trawl fleet in the North Sea to compare these pressure indicators, as the level of information used is increased. The first level is that of fleet capacity (e.g. number of vessels), the second is fishing effort, usually expressed as the number of hours fishing or days at sea, the third incorporates fishing parameters such as the proportion of time actually spent fishing, fishing speed, or gear characteristics, e.g. the size of the beam trawl in order to determine the frequency with which an area is fished, and at the fourth level, the most informative measure of fishing pressure, annual fishing mortality, is available for a few commercial species from stock assessments. For other species, it can be calculated from the lower level pressure indicators through the incorporation of the chance of individuals of a species coming into contact with the fishing gear and the encounter mortality, which is the portion of mortality caused by the passing of the gear. Comparison of trends and absolute values shows that the pressure indicators at different levels differ considerably in their description of both present and historical fishing impact in the North Sea. Therefore, for an EAFM, we advise using the highest level pressure indicator that can be obtained with the data available.
Keywords: ecosystem approach to fisheries management, fishing impact, fishing mortality, marine ecosystem, métier, pressure indicator
Received 22 February 2006; accepted 28 August 2006; advance access publication 25 October 2006.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionGeneral discussionReferences |
|---|
When implementing an ecosystem approach to fisheries management (EAFM), indicators are required to describe the pressures affecting the ecosystem, the state of the ecosystem, and the response of managers (Jennings, 2005). Such indicators can be used to support management decision-making, to track progress towards meeting management objectives, and to aid communication with non-specialist audiences (Garcia et al., 2000; Rice, 2000; Rochet and Trenkel, 2003). Many indicators have been proposed (e.g. Rice, 2000; Link, 2002; Link et al., 2002; Rochet and Trenkel, 2003), but few (if any) of those that track changes in the "state" of the marine environment or of different ecosystem components (e.g. fish, benthos, habitat) can support management directly (Rice, 2000). This is largely because the precise causes of any changes in "state" may be poorly understood, making it difficult to specify appropriate management action. To implement an EAFM successfully, therefore, it is not only necessary to have a suite of indicators that accurately and comprehensively portray the "state" of various ecosystem components, but it is also critical to have indicators that describe changes in the level of different manageable anthropogenic activities, and that indicate the impact of each activity on the various ecosystem components. Only by adequately covering both aspects will the mechanistic links between "cause" and "effect" be well enough understood to provide the advice required (Daan, 2005).
Several frameworks have been proposed for classifying environmental management indicators on this basis, e.g. the pressure state response (PSR) system (Garcia and Staples, 2000). This framework uses pressure indicators (P) to measure the pressure impacting an ecosystem component, state indicators (S) to measure the state of the ecosystem component, and response indicators (R) to measure the response of managers to the change in state. This is in line with traditional fisheries management, in which a report is made for each stock on what is considered the best state indicator (SSB, spawning-stock biomass) and best pressure indicator (F, fishing mortality), whereas the response indicator is usually the total allowable catch set by managers. As policy commitments and associated objectives generally relate to "state", for example, to conserve biodiversity, reference points, targets, or indicator trends needed to measure progress towards meeting management objectives tend most frequently to be set for "state" indicators. However, the "state" of different components of marine ecosystems can rarely, if ever, be managed directly. All managers can realistically hope to achieve is to manipulate "pressure" such that "state" indicators are kept within, or moved towards, acceptable limits. "Pressure" and "response" indicators are clearly essential in this process. Such indicators also often have the desirable properties of ease of measurement and rapid response times. Consequently, guidance for year-on-year management decision-making is often better based on pressure and response indicators, with changes in state assessed less frequently to confirm that pressure and response are affecting state as predicted (Nicholson and Jennings, 2004).
The development of pressure indicators for fisheries has in the past tended to be hampered by confusion over the difference between the actual ecological impact of fishing (mortality and habitat change) and the community level changes that are later seen as a consequence of this impact (e.g. a change in the size structure of the community). The ecosystem components for which most information is available on the direct effects of fishing are fish and benthic invertebrates (Dayton et al., 1995; Jennings and Kaiser, 1998; Hall, 1999; Collie et al., 2000; Kaiser and de Groot, 2000). Many indicators have been proposed that describe the state of these components at different hierarchical levels (e.g. population and community levels, see Frid et al., 2005; Piet and Jennings, 2005). However, the state of the individual components is not only determined by the effects of a particular anthropogenic activity such as fishing, but by a combination of all intrinsic and extrinsic factors that combine to structure populations and communities. These include a combination of biotic (e.g. competition, predation, and larval dispersal) and abiotic factors (e.g. climatically driven changes in temperature and productivity; Murawski, 1993; Clark and Frid, 2001; Kröncke, 2001). In theoretical ecology terms, impact or disturbance is the mortality caused by perturbations to the ecosystem, and fisheries impact is an anthropogenic source of mortality. Clearly, to be able to predict realistically the response of ecosystem components to fisheries impact, one must first establish the level of mortality experienced by these components before inputting this to an overall model of the factors that structure them.
In terms of indicators required to implement an EAFM, pressure indicators should ideally account for mortality to an ecosystem component that results directly from fishing, state indicators for the overall state of that component, and response indicators for the response of managers required to alter the level of pressure where there have been unacceptable changes in state. In case management objectives are set for populations of ecosystem components other than the target fish species (e.g. threatened and declining species, sensitive benthic species) or at the community level (e.g. mean weight of the fish community), the known mortality estimates of a few target fish species that come from stock assessments may not be adequate to link pressure with such state indicators. For those cases, the mortality of all ecosystem components can be calculated following the swept-area method introduced simultaneously by Pope et al. (2000) for fish and Piet et al. (2000) for benthos.
Starting from the premise that fishing mortality is the most accurate measure to describe fishing impact but acknowledging that data limitations often force us to use less-informative proxies, we introduce a framework (Figure 1) that encompasses the most common pressure indicators. The main objective of this paper is to describe this framework and how the different existing pressure indicators compare with one another, by incrementally adding information to the most basic pressure indicator, ultimately leading to indicators that describe the actual ecological impact of fishing, i.e. the level of mortality inflicted on a particular ecosystem component. This approach makes explicit the assumptions that are made at lower levels of information content and how these influence the pressure indicator.
|
We chose the Dutch beam trawl fishery in the southern North Sea as a case study. Beam trawling accounts for a high proportion of all fishing activity, particularly in the southern North Sea (Jennings et al., 1999), and the Dutch beam trawl fleet is responsible for >70% of total beam trawl effort in the North Sea. Moreover, the fishery and fleet have been intensively studied in recent years, resulting in a high level of knowledge regarding its precise operation (Rijnsdorp et al., 1998, 2000a, 2000b; Piet et al., 2000). Mortality is only estimated for the two target species of the beam trawl fishery, plaice (Pleuronectes platessa) and sole (Solea vulgaris), despite the fishery's potential to cause collateral damage to other components of the marine ecosystem. For estimates of fishing pressure on these other components, only one of the least informative measures of effort (i.e. days at sea) at a spatial scale of ICES rectangles (
30x30 nautical miles) is available. In order to exemplify how this can be improved, we used the swept-area method to estimate the mortality of a "virtual population", with specific characteristics that determine the impact the fishery has on it. The consequences of these findings for the collection of data needed to support an EAFM are discussed.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionGeneral discussionReferences |
|---|
The Dutch beam trawl fleet operates mainly in the southeastern North Sea and targets the flatfish species plaice and sole. Two principal fishing métiers are usually distinguished: "large vessels" with an engine power of 221 kW or more, and "eurocutters", with an engine power <221 kW. The two métiers differ markedly in fishing practice and gear characteristics. Typically, the large vessels deploy 2x12-m beam trawls and are not allowed to fish inside the 12 mile coastal zone or the "plaice box", whereas the eurocutters deploy 2x4-m beam trawls and are allowed to fish inside those areas.
Because beam trawling has such a high potential to cause collateral damage to other components of marine ecosystems, including fish and benthic invertebrate communities and seabed habitat, it has long been the focus of considerable scientific attention. This has led to the collection of a wide range of data, including information on the capacity, spatial coverage, and behaviour of different types of vessels within the fleet, and temporal and spatial variation in these at different scales (Rijnsdorp et al., 1998; Piet et al., 2000). Two databases that differ in their spatial resolution were analysed in this study.
- Low spatial resolution. The VIRIS database contains information on fishing activities of the entire Dutch fleet at a spatial resolution of ICES rectangles stored in individual fishers' EC logbooks. Data were extracted for the years 1994–2004. The database distinguishes different segments of the fleet based on their engine power, and contains information on the time and date of the start and end of the fishing trip, the gear used, the ICES rectangles fished, and the landings by fish species. It is designed for quota management, but is available for research purposes, and similar databases are available for other EC countries.
- High spatial resolution. The APR/VMS database consists of automated position registration (APR) and vessel monitoring through satellite (VMS) data at a resolution of 1 min latitudex2 min longitude (1x1 nautical mile) spatial units, which are referred to subsequently in text as squares. APR data are based on a sample of about 10% of the Dutch beam trawl fleet that was equipped with APR equipment for the period 1993–2000, during which the position of the vessels was recorded every 6 min (Rijnsdorp et al., 1998). The VMS data became available from 2000, when positions of all EU vessels >24 m were recorded for enforcement purposes. From September 2003, this was extended to vessels >18 m, and subsequently from the 1 January 2005 to vessels >15 m. Positions are recorded approximately every 2 h. Although the data are collected by all EC countries for enforcement purposes, not all countries have access to VMS data for research purposes. For the Dutch beam trawl fleet, VMS data from only a subset of 40% of the vessels that allowed the use of their VMS registrations are available for research purposes. In addition to detailed data on track positions, some vessels provided data on a haul-by-haul (HBH) basis of the catch of the target species, trawling speed, and times of shooting and hauling of the gear.
Level 1: Fleet capacity
The number of registered Dutch beam trawl vessels belonging to each of the two métiers based on engine-power was determined for each year from the VIRIS database.
Level 2: Fishing effort
Total annual fishing effort, in terms of the number of days at sea, was determined for each métier within the Dutch beam trawl fleet based on the VIRIS database.
Level 3: Frequency trawled
The total area of seabed swept by the Dutch beam trawl fleet in any given year (swept area, SA, in m2 y–1) was estimated from
|
| (1) |
The APR/VMS database holds registration data recorded at different time intervals. In order to combine these data sets, the total number of hours fishing per year was calculated for each set as follows:
|
| (2) |
2 h for VMS, and PD is the proportion of the fleet in the APR or VMS sample (i.e. for which data are recorded in the database). Note that the right side of this equation can be substituted directly into Equation (1). The mean trawling frequency (TF) within the area of Dutch beam trawling operations was calculated as
|
| (3) |
Level 4: Annual fishing mortality (%)
Ideally, the effect of a fishery on any ecosystem component should be expressed as an annual fishing mortality (i.e. the percentage fishing-induced number of deaths per year in relation to population abundance). In traditional fisheries science, the instantaneous fishing mortality rate F is usually calculated as part of the stock assessment process, and it can be converted easily to the more readily understood concept of annual fishing mortality through
|
| (4) |
Fishing mortality estimates of plaice and sole were available from the ICES stock assessments. For plaice, F estimates are based on both landings and discards, for sole on landings only. As the stock assessments for these species are known to have underestimated mortality in the past few years, the values for the final year were not presented.
For all non-target species for which mortality estimates cannot be obtained from the stock assessments, we use the swept-area method (Piet et al., 2000; Pope et al., 2000) and calculated AFM (%) summed over all spatial units as
|
| (5) |
In order to illustrate how encounter mortality and overlap in distribution of the species with the fishery determine the value of the indicator, we distinguish virtual populations that differ in these two characteristics. Published gear efficiency rates of benthic invertebrates lie within the range 0–90% (Bergman and van Santbrink, 2000; Collie et al., 2000). In our example, therefore, we used encounter mortality rates of 20 and 80%. For the overlap in distribution, we distinguished a species with "high overlap" (HO), which is aggregated in areas where fishing activity is highest (i.e. in spatial units where TFi is greater than the mean TF across all spatial units), and a species with "low overlap" (LO), which is aggregated in areas where fishing activity is least (TFi<mean TF). In all cases, the population with a fixed number of organisms was uniformly distributed over the spatial units that matched the criteria. More spatial units result in less organisms per unit.
We explored the same scenarios as for the Level 3 indicator: no spatial information, spatial information at low resolution (ICES rectangles), and spatial information at high resolution (1x2 min squares). For comparison of indicators across levels, we fitted linear regressions to the centred time-series (centred at year 2000), and used the slopes and intercepts to calculate the estimated value in year 2000 (intercept), and the relative change (%) over time (slopex100/intercept).
|
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionGeneral discussionReferences |
|---|
Level 1: Fleet capacity
According to the VIRIS database, the total number of registered beam trawlers declined from 378 in 1995 to 224 in 2004 (Figure 2). The reduction in the number of large vessels was much smaller in relative terms than the reduction in the number of eurocutters, so the proportion of large vessels within the fleet increased from 55% in 1995 to 63% in 2004 (Figure 2).
|
Level 2: Fishing effort
On the basis of the VIRIS database, the activity per vessel (days at sea per year) varied considerably within and between métiers (Figure 3). Overall, 87% of the large vessels spent 150–250 d at sea per year, with an average of 170 d at sea per year. For eurocutters, mean activity was much less, with an average of just 67 d at sea per year, but the distribution was skewed because 25% of eurocutters registered fewer than 10 d at sea per year, many registering only 1 d at sea per year. For both métiers, activity per vessel decreased by about 1.5 d at sea per year. Total Dutch beam trawl effort decreased from 49 765 d at sea in 1995 to 26 034 d at sea in 2004. Over the same period, the proportion of total Dutch beam trawl fishing effort undertaken by large vessels increased from 76 to 82% (Figure 2).
|
Level 3: Frequency trawled
For each of the two fishing metiers, frequency distributions of the estimated vessel speed for all APR and VMS registrations were compared with frequency distributions of recorded trawl speeds in the HBH data (Figure 4). Fishing speeds in the HBH data ranged from 3 to 6 knots for eurocutters, and from 5 to 8 knots for large vessels. APR and VMS records giving speed estimates within these ranges were considered to be "fishing" records, and the proportion of records falling into this category provided an estimate of the proportion of time spent fishing, allowing the number of hours spent fishing by each métier in an average 24 h period to be calculated (Table 1). There was little difference between the estimated fishing speeds or the proportion of time spent fishing derived from either the APR or the VMS databases (Table 1), so the values derived from the VMS database were used in Equation (1) to estimate the area of seabed swept per day by a fishing vessel belonging to each métier. Eurocutters swept an area of 1.2 km2 on average each day, and large vessels swept an area of 5.3 km2.
|
|
The VIRIS database provided information that identified all the ICES rectangles in which Dutch beam trawlers were recorded fishing in each year. Likewise, the APR/VMS database identified the fished rectangles. Knowing the area of each rectangle or square, and summing over all rectangles or squares in which fishing occurred, allowed the proportion of the total area fished annually by Dutch beam trawling operations to be estimated, depending on the spatial resolution of the data (Figure 5a). The low-resolution data indicated that just over 100 ICES rectangles, amounting to about 58% of the North Sea, was fished annually at the start of the time-series, declining to about 50% annually at the end. In contrast, the high resolution data indicated that considerably less of the North Sea was fished annually, respectively 20% (about 26 000 rectangles) and 14%. Figure 5b shows the time-series of the Level 3 indicator, the frequency fished (TF). The difference in frequency distribution between the VIRIS and APR/VMS data sets is reflected in Figure 6, which shows the occurrence of spatial units with specific trawling frequencies (per year). Frequencies above 20 per year were only observed for the high-resolution APR/VMS data. The two métiers are strongly segregated spatially, because the eurocutters, which are allowed to fish anywhere, appear to avoid the areas where the large vessels fish and concentrate in the 12-mile zone and the plaice box, where the large vessels are not permitted (Figure 7).
|
|
|
The degree to which the subset of the Dutch bottom trawling fleet for which APR/VMS data were available is representative of the entire fleet differs considerably between the period when APR data were used and that when VMS data became available (Figure 8). In the first period (before 2000), mainly large beam trawlers (15–24) and a few eurocutters (1–6) per year were included in the sample. From 2000 on, this increased to 66–143 large vessels and 17–37 eurocutters per year.
|
300 hp; and eurocutter). The number of registrations is standardized to a 1-h interval.Level 4: Annual fishing mortality
The fishing mortality (F) of plaice and sole is estimated as part of the stock assessment process (ICES, 2005) and translated into the annual fishing mortality using Equation (4) (Figure 9). The time-series of this indicator reveals that contrary to what was observed for the lower level indicators, there was no decrease in annual fishing mortality over the 10 y 1995–2004, but rather a slight (but not significant) increase of 0.04% for sole and 0.55% for plaice (Table 2).
|
|
For non-target species, there are no stock assessments, so annual fishing mortalities were estimated following the swept-area method. This resulted in time-series of annual fishing mortality for different virtual populations, with various scenarios depending on the spatial resolution of the fishing data and how the spatial resolution of the virtual population being impacted related to that of the fishery (Figure 10). The time-series of annual fishing mortality differed considerably between scenarios, both in terms of absolute value and the relative change (Figure 10, Table 2).
|
Overall, annual fishing mortality increased with increasing overlap of the population with the fishery, encounter mortality, and spatial resolution, and in the example varies between 0.7 and 80.1%. The relative change in annual fishing mortality over time also varied between scenarios, from a 1.8% decrease to a 1.8% increase.
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionGeneral discussionReferences |
|---|
We used data from the Dutch beam trawl fleet to illustrate how known pressure indicators differ in their representation of fishing impact. When more assumptions need to be made (e.g. composition of the fleet, fishing practices, or modifications to the gear) and hence the level of information content decreases, the time-series of the indicator deviates more from what we consider the best pressure indicator, annual fishing mortality, and becomes more biased. We now address some considerations regarding the implementation of these pressure indicators to support an EAFM, the interpretation of the indicator time-series, and the data required to minimize bias.
Level 1: Fleet capacity
When using fleet capacity as the pressure indicator, it is important to distinguish between métiers, where each métier is considered to have a relatively homogeneous distinct impact on the ecosystem. The fishery in a specific region can be divided into increasingly smaller and more homogeneous métiers. With the information currently available, however, it is not possible to parameterize these (sub)-métiers in terms of gear characteristics, fishing speed, and fishing area. Characterization of métiers is always arbitrary, and increasingly smaller sub-units can be created. Considering that every variation in (rigging of) the gear may alter its impact on the ecosystem, subdividing the fishery into many subdivisions (métiers) potentially allows the most accurate assessment of the ecosystem effects of the fishery. However, if effort data and information on the impact per unit effort by métier are not available, there is no point in subdividing beyond the limitations of the data.
Level 2: Fishing effort
Activity varies considerably between vessels, even within a single métier. This may be for administrative reasons, for example, when vessel quotas are transferred to other vessels of the same company. Alternatively, in the métier, where the differences in activity between vessels are largest (i.e. the eurocutters), vessels engage in different types of fishing at different times of the year, using different gears (e.g. shrimp trawl, otter trawl), and for those trips they are not registered as beam trawlers. The result of this is that measures such as decommissioning that appear to have a large effect at Level 1 may not reduce the actual pressure of a specific métier by the same proportion at Level 2. For example, a reduction in capacity of 19% of the least active eurocutters results in a reduction of just over 1% of effort in days at sea.
Level 3: Frequency trawled
For two segments of the Dutch beam trawl fleet, fishing parameters and gear characteristics were determined and used to estimate pressure at indicator Level 3, the frequency of trawling. Indicator values at this level vary between métier by more than a factor of 4, depending on gear characteristics and fishing parameter values. The data sets used to estimate the fishing parameters provided markedly different estimates of the main parameters, "trawling speed", and "proportion of the day spent fishing", which are inversely related and depend on the speed range chosen. An explanation for this discrepancy is that the calculated speed according to the APR data is underestimated because (i) the vessel does not follow a straight line between two subsequent registrations, and (ii) if the vessel starts hauling between two registrations, the calculated mean speed will decrease. These points will become increasingly important and result in an underestimation of the speed as the time interval between registrations is increased (as in the VMS data). Therefore, in order to use speed to distinguish fishing registrations from other activities that do not impact the ecosystem, it is necessary to obtain information on the activity of a vessel (e.g. HBH), and to combine this with real speed measurements (i.e. not calculated from intervals between registrations). If this needs to be based on calculated speed values, the interval needs to be set as short as possible. This approach to use speed to distinguish fishing from other activities will never be absolutely perfect, because vessels may engage in activities other than fishing at speeds within the fishing speed range, resulting in spurious fishing position registrations and an overestimation of impact.
If available, information on the spatial distribution of the fishery at the highest possible resolution needs to be incorporated when assessing the pressure on the ecosystem. We used two sources of data that differed in their spatial resolution: VIRIS data are at a relatively low resolution of ICES rectangle (30x30 nautical miles), whereas the APR/VMS data were aggregated at a relatively high resolution of 1x2 min squares (1x1 nautical mile). Comparison of the frequency distributions of these data sets shows that the resolution used makes a big difference. According to low resolution data, >50% of the North Sea is fished, of which one-third of the ICES rectangles are fished more than once a year and maximum frequency is about five times per year. According to the high resolution data, only some 20% of the North Sea is fished, of which two-thirds are fished more than once a year and the maximum frequency is up to 100 times per year. In this example, the proportion of the area fished according to the high resolution data may be slightly underestimated, because not all beam trawlers are part of the APR/VMS database available for scientific purposes. The calculated frequency was not affected by this, though, because it was weighted by a factor that raised the frequency of the sample to that of the entire fleet. The important take-home message here is that as the spatial resolution of fishing activity data increases, the proportion of the area fished declines, whereas the trawling frequency within the fished area increases on average.
Level 4: Annual fishing mortality
The annual fishing mortality differs from the lower level indicators, because it explicitly incorporates both the chance of individuals of a species coming into contact with the fishing gear and the encounter mortality, and also because it is the only indicator that describes the pressure relative to the abundance of the species. As such, it should formally be considered an impact indicator, but the distinction between pressure and impact indicators is not made in the PSR framework.
The time-series of annual fishing mortality based on stock assessments of plaice and sole show that pressure on these stocks in the 10 y 1995–2004 probably increased. This becomes even more likely when considering the systematic underestimation of mortality in the past few years by the stock assessment process (Piet and Rice, 2004). The parameters that describe the annual population mortalities of plaice and sole based on the stock assessments fall within the range observed for the different populations at high spatial resolution based on the modelled pressure indicators, suggesting that a higher spatial resolution provides more accurate estimates.
Although the increase in population mortality of plaice and sole is systematically underestimated in their stock assessments, the increase in population mortality based on the swept-area method is probably also underestimated, because we assumed encounter mortality to be stable over time, failing to take account of technical creep and increases in skipper skills which, according to Rijnsdorp et al. (2006), increase the partial fishing mortality rate in the beam trawl fishery (>221 kW) by 2.8% per year for sole and 1.6% for plaice. To what extent these values apply to non-target species, however, remains to be assessed.
When using the swept-area method to calculate the annual fishing mortality, it may theoretically vary between 0 and 100% depending on encounter mortality of that particular species–métier combination, the spatial distribution of the population relative to that of the fleet, and the resolution of the spatial distribution data. In this study, the annual fishing mortality increases with increasing encounter mortality (i.e. 20
80%), increasing overlap (i.e. lowevenhigh), and increasing spatial resolution (i.e. nslowhigh). The increase of mortality with spatial resolution is because, in the calculations, the population is distributed relative to the mean trawling frequency (TF) which, as Figure 5 shows, depends on the spatial resolution. The choice of a fixed TF, independent of the spatial resolution, would probably result in an inverse relationship with spatial resolution, because at a higher resolution, less of the area is fished, although with a higher frequency. In general, at lower spatial resolution, effort is evenly distributed over a large spatial unit. Within that unit, there will be high resolution spatial units with much higher frequencies, whereas other high resolution spatial units have lower frequencies or are not fished at all. Because of this and the fact that every haul in a heavily fished area causes less mortality than a similar haul in an area that was hardly fished or pristine, the net effect of increasing spatial resolution should be a decrease in population mortality.
The slope of the trend over time in population mortality decreases with increasing overlap: for an encounter mortality of 80% and low spatial resolution, this is –0.4% at low overlap to –1.8% at high overlap; at high spatial resolution it is +1.8% at low overlap to +0.1% at high overlap (Table 2). These large differences show that within an EAFM, target levels for mortality can only be set realistically at the highest level of information content, and that when mortality needs to be calculated (e.g. for non-target species), the spatial resolution is important. It is therefore proposed that a standard spatial resolution be used and, as long as this is not achieved, the spatial resolution should at least be reported whenever frequencies (Level 3) or calculated mortality values (Level 4) are used.
Determining the encounter mortality of a species in the gear used by a specific métier may be difficult, because almost every vessel will fish with gear with slightly different characteristics and rigging (e.g. for a beam trawl, the mesh size, the number of tickler chains, the use of a chain mat or flip-up rope), which could affect encounter mortality. It might seem appropriate therefore to distinguish each of these gear types, but the amount of work required to estimate encounter mortality would then be prohibitive. We suggest that a limited number of métiers be distinguished with standard gear types assumed for which encounter mortality can be determined.
The consequences of the above for the use of pressure indicators as part of an EAFM is that the best pressure indicator (i.e. annual fishing mortality) will differ between species, both in terms of its absolute value and its trend over time, and that this in turn may respond differently to management measures than the lower level pressure indicators. Considering this difference between species, the pressure expressed as annual fishing mortality of one or two commercial species will not be representative for the whole community. Therefore, if management objectives are set for community level indicators such as mean weight, mean maximum length, or biodiversity (Piet and Jennings, 2005), the community mortality will need to be determined as an integral of all the population mortalities that make up the community.
Here we have illustrated the process for a number of "virtual populations" of marine species, one that suffers 20% mortality in a beam trawl gear and the other 80%. Ultimately, it should be possible to calculate Level 4 indicators for real populations and communities using distribution maps of benthos and demersal fish (for example, see www.mafcons.org) and incorporating characteristic species or taxon-level information on the encounter mortality of the species making up the community in a given fished area. This is likely to be more achievable in the near future for benthic invertebrates, because distribution maps often exist for both fish and benthos, but encounter mortality has only been determined for some of the common benthic species (for review and meta-analysis, see Collie, 2000, and Kaiser et al., 2005), using sampling techniques that allow the determination of absolute pre- and post-haul abundances (see example for a limited number of benthic species in Piet et al., 2000). For fish, there are as yet no estimates of the encounter mortality caused by the passing of the gear despite the fact that in fishery science, catchability and gear efficiency are known concepts (e.g. Dickson, 1993). To further complicate the matter, encounter mortality consists not only of mortality of animals caught in the net (i.e. catchability), but also mortality through contact with the gear (e.g. after passage through the net). This is more relevant for benthos than it is for fish (Bergman and van Santbrink, 2000).
|
General discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionGeneral discussionReferences |
|---|
The usefulness in an EAFM of this framework of pressure indicators becomes apparent when the minimum level of information content required to evaluate a particular type of management measures can be identified. Effort control through decommissioning is already reflected at the lowest level, while it will show at the second level if it is implemented through a reduction in days at sea. Technical measures may show up at the third level if it involves changes in the gear characteristics that determine the area fished in relation to effort (e.g. if the width of a beam trawl is reduced), but usually only affect the encounter mortality, and will therefore only show up at Level 4. Spatial measures such as marine protected areas can only be evaluated at Level 4.
This case study demonstrates that it is possible to develop pressure indicators that describe the impact induced by fishing activities on a particular ecosystem component or system (e.g. the demersal system here), and which are appropriate for use as part of an EAFM. However, the best pressure indicators also come with extensive data requirements that at present are only marginally available even in one of the most data-rich marine environments in the world. An EAFM can only be successfully implemented and monitored if the fishing pressure can be described at a level of information content that is adequate to guide management decision-making.
|
Acknowledgements |
|---|
The work was funded by the European Commission: "European Fisheries Ecosystem Plan" (EFEP, EC project Q5RS–2001–01685) and "Managing Fisheries to Conserve Groundfish and Benthic Invertebrate Species Diversity (MAFCONS, EC project Q5RS–2002–00856). We thank Adriaan Rijnsdorp, Niels Daan, and Willem Dekker for helpful comments on an earlier draft of the paper.
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionGeneral discussionReferences |
|---|
-
Bergman M. J. N. and van Santbrink J. W. (2000) 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.
Beverton R. J. H. and Holt S. J. (1957) On the dynamics of exploited fish populations. Series II (Fisheries Investigations, London) 19:533.
Casey J. (1996) Estimating discards using selectivity data: the effects of including discard data in assessments of the demersal fisheries in the Irish Sea. Journal of Northwest Atlantic Fishery Science 19:91–102.
Clark R. A. and Frid C. L. J. (2001) Long-term changes in the North Sea ecosystem. Environmental Review 9:131–187.
Collie J. S., Hall S. J., Kaiser M. J., Poiner I. R. (2000) A quantitative analysis of fishing impacts on shelf-sea benthos. Journal of Animal Ecology 69:785–798.[CrossRef][Web of Science]
Daan N. (2005) An afterthought: ecosystem metrics and pressure indicators. ICES Journal of Marine Science 62:612–613.
Dayton P. K., Thrush S. F., Agardy M. T., Hofman R. J. (1995) Environmental effects of marine fishing. Aquatic Conservation 5:205–232.
Dickson W. (1993) Estimation of the capture efficiency of trawl gear. 1. Development of a theoretical model. Fisheries Research 16:239–253.[Medline]
Frid C. L. J., Robinson L. A., Bremner J. (2005) Muddy thinking: ecosystem based management of marine benthos. American Fisheries Society Symposium 41:651–665.
Garcia S. M. and Staples D. (2000) Sustainability indicators in marine capture fisheries: introduction to the special issue. Marine and Freshwater Research 51:381–384.[CrossRef][Web of Science]
Garcia S. M., Staples D. J., Chesson J. (2000) The FAO guidelines for the development and use of indicators of sustainable development of marine capture fisheries and an Australian example of their application. Ocean and Coastal Management 43:537–556.[CrossRef]
Hall S. J. (1999) The Effects of Fishing on Marine Ecosystems and Communities. (Blackwell Science, Oxford).
Jennings S., Alvsvaag J., Cotter A. J. R., Ehrich S., Greenstreet S. P. R., Jarre-Teichmann A., Mergardt N., Rijnsdorp A. D., Smedstadt O. (1999) Fishing effects in Northeast Atlantic shelf seas: patterns in fishing effort, diversity and community structure. 3. International trawling effort in the North Sea: an analysis of spatial and temporal trends. Fisheries Research 40:125–134.[CrossRef][Web of Science]
Jennings S. and Kaiser M. J. (1998) The effects of fishing on marine ecosystems. Advances in Marine Biology 34:201–352.[Web of Science]
Jennings S. (2005) Indicators to support an ecosystem approach to fisheries. Fish and Fisheries 6:212–332.[CrossRef][Web of Science]
Kaiser M. J. and de Groot S. J. (2000) Effects of Fishing on Non-Target Species and Habitats. (Blackwell Scientific, Oxford).
Kaiser M. J., Hinz H., Callaway R. M., Nall A., Biles C. L. (2005) Resource degradation: a subtle effect of bottom fishing. Marine Biology 146:401–408.[CrossRef]
Kröncke I. B. (2001) Review of the current knowledge on North Sea Benthos. Synthesis and New Conception of North Sea Research. SYCON no.12., Zentrum für Meeres- und Klimaforschung der Universität Hamburg.
Laurec A., Biseau A., Charuau A. (1991) Modelling technical interactions. ICES Marine Science Symposium 193:225–236.
Link J. S. (2002) What does ecosystem-based fisheries management mean? Fisheries 27:18–21.
Link J. S., Brodziak J. K. T., Edwards S. F., Overholtz W. J., Mountain D., Jossi J. W., Smith T. D., Fogarty M. J. (2002) Marine ecosystem assessment in a fisheries management context. Canadian Journal of Fisheries and Aquatic Sciences 59:1429–1440.[CrossRef]
Murawski S. (1993) Climate change and marine fish distributions: forecasting from historical analogy. Transactions of the American Fisheries Society 122:647–658.[CrossRef]
Nicholson M. D. and Jennings S. (2004) Testing candidate indicators to support ecosystem-based management: the power of monitoring surveys to detect temporal trends in fish community metrics. ICES Journal of Marine Science 61:35–42.[CrossRef][Web of Science]
Piet G. J. and Jennings S. (2005) Response of potential fish community indicators to fishing. ICES Journal of Marine Science 62:214–225.
Piet G. J. and Rice J. C. (2004) Performance of precautionary reference points in providing management advice on North Sea stocks. ICES Journal of Marine Science 61:1305–1312.[CrossRef]
Piet G. J., Rijnsdorp A. D., Bergman M. J. N., van Santbrink J. W., Craeymeersch J., Buijs J. (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.
Pope J. G., Macdonald D. S., Daan N., Reynolds J. D., Jennings S. (2000) Gauging the impact of fishing mortality on non-target species. ICES Journal of Marine Science 57:689–696.
Rice J. C. (2000) Evaluating fishery impacts using metrics of community structure. ICES Journal of Marine Science 57:682–688.
Ricker W. E. (1975) Computation and interpretation of biological statistics of fish populations. Bulletin of the Fisheries Research Board of Canada 191:.
Rijnsdorp A. D., Buys A. M., Storbeck F., Visser E. G. (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.
Rijnsdorp A. D., Daan N., Dekker W. (2006) Partial fishing mortality per fishing trip: a useful indicator of effective fishing effort in mixed demersal fisheries. ICES Journal of Marine Science 63:556–566.
Rijnsdorp A. D., Dol W., Hoyer M., Pastoors M. A. (2000a) Effects of fishing power and competitive interactions among vessels on the effort allocation on the trip level of the Dutch beam trawl fleet. ICES Journal of Marine Science 57:927–937.
Rijnsdorp A. D., van Maurik Broekman P. L., Visser E. G. (2000b) Competitive interactions among beam trawlers exploiting local patches of flatfish in the North Sea. ICES Journal of Marine Science 57:894–902.
Rochet M.-J. and Trenkel V. M. (2003) Which community indicators can measure the impact of fishing? A review and proposals. Canadian Journal of Fisheries and Aquatic Sciences 60:86–99.[CrossRef]
Wardle C. S. (1988) Understanding fish behaviour can lead to more selective fishing gears. Proceedings of the World Symposium on Fishing Gear and Fishing Vessel Design. (The Newfoundland and Labrador Institute of Fisheries and Marine Technology, St John's, Newfoundland, Canada)12–18.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  G. J. Piet, R. van Hal, and S. P. R. Greenstreet Modelling the direct impact of bottom trawling on the North Sea fish community to derive estimates of fishing mortality for non-target fish species ICES J. Mar. Sci., October 1, 2009; 66(9): 1985 - 1998. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. P. R. Greenstreet, G. J. Holland, T. W. K. Fraser, and V. J. Allen Modelling demersal fishing effort based on landings and days absence from port, to generate indicators of "activity" ICES J. Mar. Sci., June 1, 2009; 66(5): 886 - 901. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. P. R. Greenstreet, H. M. Fraser, and G. J. Piet Using MPAs to address regional-scale ecological objectives in the North Sea: modelling the effects of fishing effort displacement ICES J. Mar. Sci., January 1, 2009; 66(1): 90 - 100. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. Pedersen, H. Fock, J. Krause, C. Pusch, A. L. Sell, U. Bottcher, S. I. Rogers, M. Skold, H. Skov, M. Podolska, et al. Natura 2000 sites and fisheries in German offshore waters ICES J. Mar. Sci., January 1, 2009; 66(1): 155 - 169. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. M. Trenkel, M.-J. Rochet, and B. Mesnil From model-based prescriptive advice to indicator-based interactive advice ICES J. Mar. Sci., May 1, 2007; 64(4): 768 - 774. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||