Reconciling Fisheries with Conservation Nielsen, Dodson, Friedland, Hamon, Musick, and Verspoor Volume I

ISSN 0892-2284 ISBN 978-1-888569-80-3

Reconciling Fisheries with Conservation: Proceedings of the Fourth World Fisheries Congress

American Fisheries Society Symposium 49 Jennifer Nielsen, Julian J. Dodson, Kevin Friedland, Troy R. Hamon, Jack Musick, and Eric Verspoor, editors

American Fisheries Society Symposium 49:1019–1039 © 2008 by the American Fisheries Society

Ecosystem Modeling Approaches for South African Fisheries Management LYNNE J. SHANNON* Marine and Coastal Management, Offshore Resources Private Bag X2, Rogge Bay 8012, South Africa

COLEEN L. MOLONEY University of Cape Town, Department of Zoology Rondebosch 7701, South Africa

PHILIPPE M. CURY Institut de Recherche pour le Développement, Centre de Recherche Halieutique Méditerranéenne et Tropicale Avenue Jean Monnet, BP 171, Sète Cedex 34203, France

CARL D.

VAN DER

LINGEN

AND

ROBERT J. M. CRAWFORD

Marine and Coastal Management, Offshore Resources Private Bag X2, Rogge Bay 8012 South Africa

PIERRE FRÉON Institut de Recherche pour le Développement, Centre de Recherche Halieutique Méditerranéenne et Tropicale Avenue Jean Monnet, BP 171, Sète Cedex 34203, France

KEVERN L. COCHRANE Food and Agriculture Organization of the United Nations, Fisheries Resource Division Via della Terme di Caracalla, Rome 00100, Italy

Abstract.—A variety of ecosystem modeling approaches is available, some of which have been successfully applied to the southern Benguela ecosystem. How these models could be useful for fisheries management and how they might assist in reconciling fisheries with conservation, has not always been clear. At the end of 2002, a workshop was held to introduce the concept of an ecosystem approach to fisheries (EAF) management to South African scientists and managers. It was recommended that an EAF be implemented as an incremental process with immediate effect in South Africa. An ecosystem modeling perspective should be incorporated into existing single species management recommendations by using ecosystem models to test the implications of these recommendations. This paper describes the rich variety of ecosystem models that are being applied to the southern Benguela. Focusing on the South African small pelagic fisheries, it documents how these models are assisting the current implementation of an EAF in the southern Benguela and suggests how future management and conservation issues should be addressed in South African fisheries. *

Corresponding author: [email protected]

1019

SHANNON ET AL.

1020

Introduction There has been a global paradigm shift to move towards an ecosystem approach to fisheries (EAF) management. Most of the principles of an EAF are contained in a number of binding or voluntary arrangements, agreements, conventions and codes (S. Garcia, Food and Agriculture Organization of the United Nations, personal communication). These instruments started with the 1982 Convention on the Law of the Sea and the 1995 FAO Code of Conduct for Responsible Fisheries. More recently, in 2001, the Reykjavik Declaration insisted on incorporating ecosystem considerations into fisheries management. This was reinforced and a move towards EAF was encouraged before 2010 at the World Summit of Sustainable Development (WSSD) in Johannesburg in 2002. Nonetheless, operational fisheries management is still largely on a single-species basis. The challenge facing nations worldwide is how to incorporate ecosystem considerations in fisheries management. One of the targets agreed upon at the WSSD, held in Johannes-burg in 2002, reads as follows: “[to] Encourage the application by 2010 of the ecosystem approach, noting the Reykjavik Declaration on Responsible Fisheries in the marine Ecosystem and decision 5/6 of the Conference of Parties to the Convention on Biological Diversity” (Chapter 4, paragraph 29). In South Africa, party to this agreement, work is underway to formalize the inclusion of results from ecosystem models and studies in our fisheries management advice. A workshop was convened in Cape Town in December 2002 to introduce and examine the options for implementing an EAF in South Africa. Several available modeling tools of potential use in ecosystem-based fisheries management were considered (see Shannon et al. 2004a for an overview). It was recommended that an EAF be implemented in South Africa as an incremental procedure with immediate

effect. The envisaged starting point would be to use ecosystem models to provide guidance on reference points and broader management objectives currently set according to singlespecies assessments (Shannon et al. 2004a). In mid-2003, a special EAF Task Group was established by the Marine and Coastal Management branch of the South African Department of Environmental Affairs and Tourism. Terms of reference for this group have been agreed upon and the group has been tasked with coordinating the process of implementing an EAF in South African fisheries. As a first step, options for inclusion in the management of the pelagic fishery are being considered. Given the substantial multidisciplinary research that has been undertaken with respect to South African pelagic fish stocks and their dynamics (Moloney et al. 2004), the pelagic fishery was selected as the most suitable starting point. To continue the process towards an EAF in South Africa, a policy document on ecosystem approaches for South African fisheries management is required. The policy document would serve as a set of guidelines for incorporating ecosystem considerations into the management of the various South African fishing sectors. Preparation of a document such as this will require inputs and buy-ins from a wide range of stakeholders (industry, conservation bodies, scientific bodies, etc.). In addition to this national initiative, a regional EAF project was launched in southern African in 2004, under the auspices of the Benguela Current Large Marine Ecosystem (BCLME, see web site www.bclme.org). The BCLME project on ecosystem approaches for fisheries (EAF) management in the BCLME is investigating the feasibility of EAF management in the BCLME region that covers Angola, Namibia, and South Africa and is addressing EAF from both national and regional perspectives.

BENGUELA ECOSYSTEM MODELING

Science, in its broadest sense, is central to the project and will inform the stakeholder group on the feasibility and consequences of obtaining ecosystem objectives. The approach to be followed in the BCLME project is based on that which was used in developing management procedures for some of the single-species fisheries such as those for hakes, rock lobster, anchovy, and sardine (Cochrane et al. 1998; Geromont et al. 1999). It is anticipated that consultation on objectives and management approaches will move forward in an iterative manner based on scientific evaluation of the consequences of these options for the target species, other species, the ecosystem as a whole, and the social and economic outputs of the fishery. This evaluation will require a range of suitable models, including single-species, multispecies, ecosystem, social and economic models, and combinations of them. Fortunately, South Africa has well developed assessment methods for many key species and, in addition, considerable experience in multispecies and ecosystem models. Nevertheless, uncertainty will be a major factor and can be explored using models. Approaches that are robust to the uncertainties will be essential to ensuring the provision of reliable scientific advice. In this paper we present the ecosystem (sensu lato) models presently available for the southern Benguela region, and, using the small pelagic fishery as a first case, examine how these models may help in progression towards an ecosystem approach to fisheries.

Ecosystem Models for the Southern Benguela Ecosystem For the South African situation, a number of ecosystem or multispecies modeling approaches have been developed, each having its own objective(s) as well as limitations, but all adding to the pool of available knowledge from which fisheries managers may draw advice for an EAF. Several of these have been documented

1021

in Shannon et al. (2004b) and elsewhere. Here we highlight a selection of the ecosystem and multispecies models applied in the southern Benguela ecosystem. Mass Balance Approach (Ecopath with Ecosim)

Ecopath with Ecosim (EwE) (Christensen and Pauly 1992; Walters et al. 1997) is a massbalance trophic modeling approach that has been widely applied in marine ecosystems. It provides a standardized modeling approach to analyze food webs, facilitating meaningful comparisons to be made between ecosystems (e.g., Moloney et al. 2005). Ecopath with Ecosim models have been used to examine the trophic structure, functioning and trophic effects of fishing in the southern Benguela upwelling ecosystem (e.g., Jarre-Teichmann et al. 1998; Shannon 2001; Shannon et al. 2003, 2004c; Shannon and Moloney 2004; Figure 1). These models have been useful in identifying data gaps and important, sensitive interactions and have assisted in prioritizing research (Shannon et al. 2004a). The user-friendly framework of EwE enables a user with limited mathematical and programming skills to construct a useful ecosystem model, which they can then analyze on the basis of intelligent and informed judgment (Cochrane 2001). Many of the same risks pertaining to singlespecies models are also associated with EwE, such as inaccurate estimates of biomass, misinterpretation of trends in data and accounting for the effects of environmental changes interacting with effects of fishing (Christensen and Walters 2004). Additional shortcomings may affect EwE, such as unexpected functional responses of predators to their prey, diet selection, competition among predators, the influence of regime shifts, disease- and parasite- transmission, parasitism and mutualism (reviewed in Cury et al. 2005a). As with any modeling approach, underlying assumptions

1022

SHANNON ET AL.

Figure 1. Example of EwE simulation to explore possible directional changes in ecosystem components under an altered exploitation strategy: Catches of various groups in the southern Benguela ecosystem, simulated over 50 years when modeled fishing mortality rate of anchovy is doubled between year 10 and year 15. Catches expressed as proportion of mean catches in the 1980s (Shannon 2001).

require careful consideration. Plaganyi and Butterworth (2004) envisage that prudent applications of EwE may contribute to the extension of single-species operational management procedures (OMPs) to include multispecies aspects, emphasizing the value of considering variability about predicted ecosystem responses. Size-based approaches

By means of object-oriented modeling, a size-based approach has been applied in the ecosystem context (Shin and Cury 2001). This spatially-disaggregated, multispecies

model (called Osmose) considers predation as a size-based, opportunistic process dependent on the spatial overlap of fish predators and suitably-sized fish prey. Predation, growth, reproduction, and mortality are modeled at the level of the fish groups in a hierarchical structure, considering cohorts and species. Shin et al. (2004) applied the Osmose model to a suite of 12 fish species in the southern Benguela to explore the ecosystem effects of various hypothetical fisheries scenarios (Figure 2). Available historical data on sizes of fish caught can be used in the ecosystem con-

1023

BENGUELA ECOSYSTEM MODELING

1.4 7

1.2

8 4 2

B/Bref

1 0.8

13

0.6

9 5

0.4 6

11 0.2 0

10 1

3 12

1 1.5

2

2.5

3

Fsmall pelagic multiplier

3.5

4

Figure 2. Osmose simulation of effects on species biomass of an increase in fishing mortality of small pelagic fish (anchovy, sardine, round herring). Biomass is plotted relative to its original biomass of reference (B/Bref) against a multiplier of the fishing mortality rate applied to small pelagic species (Fsmall multiplier). Species are indicated by number 1: anchovy, 2: chub mackerel, 3: shallow-water pelagic Cape hake, 4: deep-water Cape hake, 5: horse mackerel, 6: kingklip, 7: lanternfish, 8: lightfish, 9: round herring, 10: sardine, 11: silver kob, 12: snoek, 13: all species (Shin et al. 2004).

text. Yemane et al. (2004) have examined linefish catch, effort, and size distributions in four areas off South Africa. By means of size spectra, they have been able to track the relative declines in the sizes of fish caught in the line fishery as fishing pressure has increased over time. Further, dominance curves reflecting the changing distribution of biomass among species have proved a useful mean of quantifying the ecosystem effects of fishing in different geographical areas off the South African coast (Yemane et al. 2004).

Individual Based Models (IBMs) of Early Life History

Individual based models can supplement traditional population models by providing insights about how ecological factors that affect individual fish manifest at the population level. They can be coupled to 3D hydrodynamic models to track the trajectories of passive particles, a common approach in recent studies of early stages of fish (Werner et al. 1996; Heath and Gallego 1997; Hinckley et al. 2001). IBMs are computer intensive, and can generate a large amount of information

1024

SHANNON ET AL.

that is difficult to interpret. It can also be difficult to test the validity of their results. However, with careful planning and judicious use, they can be used to test hypotheses about the interactions of individual fish with their environment and fish behavior. One of the main advantages of IBMs is that they capture variability among individuals and how this variability might affect population-level dynamics; population models that assume all individuals are identical are unable to do this effectively. In diverse and changing environments, such as are found in the shelf areas of the southern Benguela, it is important to assess the role of a constantly changing ocean on fish population dynamics. IBMs have been coupled to a regional southern Benguela ocean model that generates flow, temperature, and density fields (Penven et al. 2001), allowing investigations of the effects of spatial and temporal variability in the environment on recruitment. A series of IBMs has been applied to the early life history stages of anchovy (Huggett et al. 2003; Mullon et al. 2002; Mullon et al. 2003; Parada 2003) and sardine (Miller et al. 2006) in the southern Benguela, where recruitment is influenced by location and timing of spawning (Huggett et al. 2003), egg buoyancy and spawning depth (Parada et al. 2003), temperature and vertical migration of larvae (Mullon et al. 2003), and by interannual variability in the temporal and spatial patterns of key environmental variables. At present, these models are only exploratory tools aimed at investigating processes (Figure 3). The next (ongoing) steps are to couple IBMs to biogeochemical models in order to take into account the effect of plankton abundance and availability on larval growth and survival. Finally, short-term prediction of prerecruitment strength according to environmental conditions during early life history are intended, although preliminary results suggest that there are so many interacting factors

responsible for recruitment success (including during the juvenile stage) that our capacity for predicting recruitment might be limited. Whereas the IBMs described above are not strictly ecosystem models since they focus on a single species (anchovy) only, their multidisciplinary nature (coupling biological and environmental processes) can be seen as a progression from single species models. Viability Modeling Approach

The viability approach has received mathematical treatment, which is known as the “viability theory1” (sensu Aubin 1991; Aubin 1996). Viability models do not lead to optimizing a time-related criterion as in optimal control theory but instead define all viable evolutions of a dynamic system under uncertainty. At each time the viability model satisfies specified constraints on the state or dynamics of the systems. This approach is an attempt to describe possible evolutions of a dynamic system, the results providing viable trajectories regarding key components of the ecosystem. These trajectories are defined within the viability kernel, which contains all the viable trajectories and enables us to understand what is possible and impossible in terms of the ecosystem constraints. Thus this type of model can integrate constraints of all kinds (conservation issues, objectives in fisheries, and ecological constraints) and simultaneously considers multiple management objectives. Objectives are phrased as limit reference points, indicating ecosystem states that should be avoided, as opposed to target reference points to be reached. Mullon et al. (2004) have recently applied the viability approach using the southern 1 It is important to note that it has nothing to do with the population viability analysis that is a tool used in conservation biology for assessing the effectiveness of different management options for threatened species.

BENGUELA ECOSYSTEM MODELING

1025

Figure 3. IBM model of anchovy early life history stages, showing (a) an example of the surface structure of the temperature and currents (arrows, only drawn every third vector) in the southern Benguela region simulated by a 3D hydrodynamic model (known as PLUME) to which IBMs are coupled (Penven et al. 2001); (b) output from an IBM experimental simulation showing the position of individual particles in the 3D hydrodynamic model (Huggett et al. 2003); results from an IBM showing transport success of particles representing anchovy eggs and early larvae as a function of (c) spawning area (the western Agulhas Bank [WAB], and the inshore [in] and offshore [off] regions of the Central (CAB) and Eastern Agulhas Banks [EAB]); and (d) a comparison between modeled transport success and observed patterns in anchovy egg abundance from three ichthyoplankton survey programmes (from van der Lingen and Huggett 2003).

Benguela ecosystem as a case study (Figure 4). They considered five components that interact trophically as follows, detritus, phytoplankton, zooplankton, pelagic fish, and demersal fish. Fisheries impacts in the southern Benguela test case were modeled and results suggested that

distributing fishing pressure across both pelagic and demersal fisheries was more sustainable than restricting exploitation to a similar level of fishing (i.e., same combined catch) on either pelagic or demersal fish communities alone (Mullon et al. 2004).

SHANNON ET AL.

1026

60 Pel 40

20

0

30

20 Dem

10

0

0 50 100 Zoo

150 200

Figure 4. Example of a 3D slice of a viability kernel approximation for the southern Benguela. The axes correspond to zooplankton biomass (horizontal), pelagic fish biomass (depth) and demersal fish biomass (vertical) (from Mullon et al. 2004).

It is recognized that there is a need for an EAF to consider dynamical interactions between key components of the ecosystem, and that an Ncomponent approach and graphical representation of these dynamics are required. Viability theory has been developed since the 1980s but has received little attention in fisheries sciences and management, while ultimately it may contribute to conciliate different management approaches as well as different management objectives (Cury et al. 2005b). Simultaneous application of the viability approach to different ecosystems will be useful to develop this new ecosystem modeling approach. However, there are still mathematical and computational limitations to calculation of the viability kernel for six or more interacting ecosystem components. Geographic Information Systems

Attention to spatial aspects of fisheries will be important for an EAF. A geographic informa-

tion system (GIS) that assembled all available environmental, survey, and catch data, as well as information collected from transmitters to satellites, has been developed to quantify spatial and temporal distributions of important fish species, top predators, and fisheries in the southern Benguela ecosystem (Drapeau et al. 2004; Pecquerie et al. 2004; Fréon et al. 2005). Density distributions were mapped for 15 fish species by combining six databases on commercial catches and scientific surveys (Pecquerie et al. 2004). Areas exploited by the pelagic and demersal fishing fleets were also mapped, as well as foraging areas of a marine mammal (seal) and two marine bird (gannet and penguin) predators estimated from individual satellite or GPS tracking data (Fréon et al. 2005). Drapeau et al. (2004) developed three spatial indicators to quantify potential interactions between any couple of species in space and time. Fréon et al. (2005) proposed seven specialized ecosystem indicators to quan-

BENGUELA ECOSYSTEM MODELING

tify trophic interactions between the pelagic and demersal fisheries, top predators (foraging areas), and fish. These were 1) biodiversity, 2) connectivity, 3) mean ratio between the exploited (fishing) area and the distribution area per species, 4) exploited fraction of the ecosystem, 5) the total catch per exploited area per fishery, 6) mean bottom depth of catches, and 7) mean distance from the coast of catches. The spatialized biodiversity indicator is associated with a map that points toward areas that deserve attention for protection of species, for instance through the development of marine protected areas (MPAs). The connectivity indicator reflects the average overlap (Figure 5) between the areas occupied by or foraged in by fish species and their predators, including fishers. A high value of this indicator underlines the importance of EAF because the effect of exploitation will propagate to many other species and fisheries. The last five indicators allow spatial changes in fishing pressure to be tracked over time. Pressure indicators such as those derived from GIS are information-rich, easier to interpret, quicker to compute and less prone to bias than conventional indices such as standardized fishing effort or the global CPUE. However, because of this simplicity, their aim is not to

1027

replace conventional indicators simply because crucial factors such as the changes in fishing unit capacity and technological improvements are not taken into account. These indicators may be of particular assistance in setting up and assessing the usefulness of marine protected areas (MPAs) for fisheries and ecosystem management, would be useful for assessing the observed and likely ecosystem impacts of fishing, and would provide a quantitative means of feeding ecosystem considerations into fisheries management. Minimum Realistic Models

Punt and Butterworth (1995) used a minimum realistic modeling (MRM) approach to explore the interactions between seals, Cape hake, and the Cape hake-directed fishery on the west coast of South Africa (Figure 6). Minimum realistic models are formally fitted to available data on abundance and diets, taking some account of second order effects, in this case density-dependent natural mortality of Cape hakes resulting from intrageneric or intraspecific cannibalism. Simulations using this seal-hake MRM suggested that culling seals, which prey heavily on hake and therefore evoked conflict with hake fishers, would have minimal, if not detri-

Figure 5. Cumulated Relative Overlapping Indices (ROA) derived using a GIS for the major species and fleets of the southern Benguela ecosystem (from Fréon et al. 2005).

1028

SHANNON ET AL.

Figure 6. Schematic diagram showing the relationships modeled by Punt and Butterworth (1995) using a hierarchy of operating models (,minimum realistic modeling approach).

mental effects on the hake fishery (Punt and Butterworth 1995). The MRM approach focuses on a few species of particular interest, and ignores other ecosystem components considered to have minimal impact on the sub-system being examined; similar difficulties are experienced to those encountered by EwE in deciding how much ecological detail to include in the model. Mori and Butterworth (2004) highlight the importance of long-term monitoring of biological parameters (e.g., age-at maturity) required for these kinds of multispecies models. One-Way Interaction Models Between Predators and Their Prey

In the Benguela system, it is thought that competition for food with purse-seine fisheries (catching mainly sardine and anchovy) has led to severe decreases and an unfavorable conservation status of some seabirds (Crawford 1998, 1999). It is necessary to manage the

fisheries in such a manner as to conserve the seabird populations (Crawford 2004). The shortage of food for seabirds has led to reduced recruitment of birds to many of the breeding populations (e.g., Crawford et al. 1999; Crawford 2003), although in cases of severe food shortage there may also be mortality of adult birds (e.g., Crawford et al. 1992a). Models of seabird populations have been coupled to models of their fish prey using functional relationships (Figure 7) that have been empirically determined (e.g., between the proportion of the adult population of the seabird that breeds, or number of chicks fledged per pair) in a given year and the biomass of fish prey available in that year (Crawford et al. 1992b). These models enable the long-term impact on seabirds of different fishing strategies to be explored. Off South Africa, seabirds account for only a small proportion of the overall mortality of pelagic fish (8%, Shannon et al. 2003), of which a substantial amount is taken by migratory sea-

BENGUELA ECOSYSTEM MODELING

1029

Figure 7. Relationship between the combined biomass of anchovy and sardine and the number of swift terns breeding at localities in South Africa’s western Cape (plotted from information in Crawford 2003).

birds (Crawford et al. 1991). Therefore, at the present level of abundance of seabirds, a oneway interaction model, in which fluctuations in prey abundance impact on the predator population but not vice versa (Butterworth and Thomson 1995), is considered adequate. For seabirds of conservation concern, because their populations have been much reduced, it is necessary to have some target for recovery. Stochastic modeling, drawing demographic parameters from their observed distributions (e.g., Shannon and Crawford 1999), has been used to assess the probability of extinction of seabird colonies of different sizes (e.g., Crawford et al. 2001), and hence to determine target population levels. Criteria proposed by IUCN (The World Conservation Union) for identifying the conservation status of a species are useful in identifying risks of extinction that are to be avoided, e.g. a species is classed as “Vulnerable” if the likelihood of extinction is greater than 10% within 100

years (http://www.redlist.org/info/ categories_criteria. html).

Management of Pelagic Fish as Part of the Southern Benguela Ecosystem The Role of Small Pelagic Fish in the Southern Benguela

In recent years anchovy and sardine stocks in the southern Benguela have shown marked increases; anchovy stock size being characterized by a sudden and sustained high recruitment observed since 2000 whereas the sardine stock has grown steadily, particularly since the late-1990s (Figure 8). These increases have had a positive impact on the pelagic fishery, with landings (>90% of which are anchovy and sardine) exceeding 500,000 mt annually for the period 2001–2003 (landings were 445,000 mt in 2000). These are the fourth, fifth, and sixth years that pelagic landings have

SHANNON ET AL.

1030

3

3

Anchovy

2 1

1

0

0 3

Sardine

2

Sardine

2

1

1

0

0 3

Chokka squid

2

Gannet breeders

2

1

1

0

0 3

Large horse mackerel

2

Seal pups

2

1

1

0

0 3

Snoek

2

2

1

1

0

0 3

Hake

2

Exploitable M. paradoxus west coast

Exploitable M. capensis south coast

2

1 0

Anchovy

2

1 1980

1985

1990

Year

1995

2000

0

1980

1985

1990

Year

1995

2000

Figure 8. Time series of catches (left panel) and abundances (right panel) of selected groups (important fished stocks or species with conservation implications) in the southern Benguela, plotted as a proportion of the mean catch or abundance estimated per species (averaged for the years for which data or estimates were available).

exceeded half a million tons since 1949 (Marine and Coastal Management unpublished data), and, together with fishery-independent

survey data (van der Lingen and Huggett 2003; J. Coetzee, J. Rademan, D. Merkle, and N. Twatwa, Marine and Coastal Management,

BENGUELA ECOSYSTEM MODELING

personal communication), indicate that small pelagic fish populations in the southern Benguela are large and at levels not seen during the past 20 years. The southern Benguela upwelling system appears to be largely wasp-waist controlled; zooplankton are controlled from the topdown by these small pelagic fish, which also control their predators from the bottom-up (Cury et al. 2000). Anchovy and sardine are prey of Cape hakes, sustaining the commercially most valuable fishery in the southern Benguela. The large pelagic fish, snoek, is an important species in both the recreational fishery and the commercial line fishery and preys on smaller fish, in particular anchovy and sardine. Using GIS indicators, contrasting overlaps in space and time were found between hake (predator) and small pelagic fish (prey), and also between the predatory large pelagic fish snoek, and its small pelagic fish prey (Pecquerie et al. 2004; Drapeau et al. 2004; Fréon et al. 2005). The feeding area of breeding African penguins, a species of conservation concern, on the west coast of South Africa is totally overlapped by the exploitation area of the pelagic fleet (Fernández Moreno 2003; Fréon et al. 2005). For other top predators such as gannets or Cape fur seal, the overlapping is less severe. When fitting EwE models to available time series of catches and abundances, Shannon et al. (2004c) found improved fits when model parameterization that was in line with waspwaist control was adopted (when compared to parameterization under other combinations of top-down or bottom-up control). Fishing has been found not to be the main or sole cause of stock fluctuations in the southern Benguela ecosystem (Shannon et al. 2000; Shannon et al. 2004c), underlining the importance of understanding the mechanisms and interactions involved in the functioning of this system.

1031

Management Procedure for Small Pelagic Fish in the Southern Benguela

Important interactions between small pelagic fish and fisheries and other fish and fisheries should be identified and considered when managing individual fisheries. Management decisions made in one fishery may well have implications for other fisheries, pointing to the importance of simultaneously considering and managing different fisheries under some kind of overarching and interactive framework. A management procedure is a set of rules that are preagreed by scientists, industry, managers and decision makers, and their advisers, to guide the use of fishery data for the regulation of fisheries. Management procedures are in place for three South African fisheries, namely the pelagic, hake, and west coast rock lobster fisheries (Cochrane et al. 1998; de Oliveira et al. 1998; de Oliveira 2003). Operational management procedures (OMPs) used to manage South Africa’s pelagic fishery have evolved considerably over the last decade in response to increasing knowledge of the dynamics of the two target species and the behavior of the fisheries, and they continue to be refined. Decadal-scale fluctuations in pelagic fish stock sizes pose an enormous challenge to fishery managers. Management of South Africa’s pelagic fishery is further complicated by mixed shoaling of juvenile sardine with anchovy, which results in a bycatch of juvenile sardine taken in anchovy-directed fishing operations. This operational interaction necessitates that the two species are managed together, and that some trade-off between them (in terms of catch levels and risk of resource collapse) is required since simultaneous optimization of anchovy and directed sardine catches is not possible. Management procedures employed since 1994 have recognized this trade-off.

1032

SHANNON ET AL.

However, despite the commitment by the regulatory authority to incorporating ecosystem approaches into management of South Africa’s pelagic fishery (see above) and substantial discussions as to how this may best be achieved, the lack of an agreed-upon approach or method for such incorporation between stakeholders has hampered efforts in this regard to date. Given that OMPs have a strong requirement for buyin by stakeholders, particularly industry, deriving mutually-agreed upon strategies is a necessity. Considerations for incorporation of EAF into the OMP used to manage the pelagic fishery could include some of the following: • The determination through ecosystem modeling of target reference points and limit reference points (either as absolute biomass levels, proportions of stock size, or some combination of both) that address the “ecosystem” requirements (principally forage requirements by predators) for small pelagic fish; • Currently, candidate OMPs for the pelagic fishery are assessed through simulations of anchovy and sardine populations that produce 20-year projections based on population assessment models, and output performance statistics such as average annual catch, interannual catch variability, and bycatch levels under varied fishing strategies (de Oliveira 2003). An assessment of how incorporation of ecosystem considerations into the OMP could impact on pelagic fish stock sizes and catch levels could be obtained from running simulations under scenarios that would include a range of ecosystem parameters (such as minimum biomass levels or the temporary closure of particular areas) and different fishing strategies, and to compare the performance statistics arising from those simulations to candidates not explicitly accounting for ecosystem considerations; • Establishing minimum biomass levels (for anchovy and sardine or a combination of both)

that could act as a “reservoir” for the ecosystem, below which “exceptional circumstances” would be declared, with an appropriate change in management measures. Under “exceptional circumstances,” deviations from the OMP are permitted such that the normal constraints on total allowable catch (TAC) levels (such as maximum interannual reductions in TAC) do not apply and reduced or zero catch levels may be recommended. Minimum, species-specific biomass levels for the declaration of “exceptional circumstances” already exist in the current OMP, although these levels were selected primarily on considerations pertinent to the species themselves (i.e., the likelihood of the stock recovering from low biomass levels) and not ecosystem considerations per se. Incorporation of ecosystem considerations would result in minimum biomass levels that are higher than current levels; • Temporal exclusion of the pelagic fishery from limited coastal areas around some densely populated penguin colonies during the breeding season, especially in periods of low pelagic stock abundance; • Use of functional relationships that link models of predators and their prey to ensure sufficient escapement of fish to maintain or grow populations of predators of conservation concern (particularly some seabirds, see section above); • Consideration to be given to refining estimates of natural mortality rate (M, predation) used in the OMP. M is currently assumed as constant for all years, whereas use of a variable annual natural mortality rate (selected from a distribution of values derived from ecosystem models) could be explored. Implementation of an EAF will require that management objectives are carefully negotiated and defined for each exploited and

BENGUELA ECOSYSTEM MODELING

nonconsumptively exploited resource. Subsequently, what is required at the overarching scale is a process for implementing the agreed policy objectives and a framework through which the implementation process can be informed by appropriate research and monitoring activities. For example, Shannon and Moloney (2004) present guidelines for the establishment of a framework of research and other activities to support an EAF in South Africa. Taking this a step further, they propose the extension of their four-step framework to a formal management procedure for ecosystems, which would simultaneously consider important marine resources (fished, nonconsumptively exploited, and those of conservation concern). Ecosystem models and analyses should be updated continuously and the ecosystem management procedure should be revised on a regular basis. This approach would need careful consideration and testing but may have potential for application to South African fisheries.

The Model Toolbox/Ecoscope Single-species fisheries models fail to capture two important aspects of the ecology of the species being modeled: their interactions (especially feeding interactions) with other species and their spatial distributions. These limitations mean that some potential management objectives cannot be addressed adequately by such single species models. For example, reliable information on trophic interactions is required in order to address management objectives that aim to exploit or conserve species that interact strongly with the species in question. Furthermore, information on spatial distributions is required in order to use closed areas as an effective management tool. These are two examples of how the suite of models described here can supplement singles-species models (Figure 9).

1033

Modeling complex systems and predicting ecological trajectories that are important for EAF are challenging tasks and uncertainty remains an important constraint. However, the recent and rapid scientific developments in modeling demonstrate that scientific expertise can substantiate an EAF. Models can be the converging point of assessing our human actions on ecosystems while ecosystem-based indicators can build bridges between science and management. It is clear that the most effective progress towards an EAF will be to tackle it from a multifaceted perspective. Starfield et al. (1988) termed this the “the toolbox approach” whereby a variety of decision-driven models are used to tackle resource management problems, depending on the management objectives defined. He likened the use of models to that of carpenter’s tools, each model being most useful for the specific purpose for which it was developed. This is exactly in line with the “ecoscope” proposed by Ulanowicz (1993), expanded upon by Cury (2004) as a means of assembling the operational tools we need to define clear long term objectives, integrate knowledge around key social objectives and ecological constraints, and mobilizes our efforts. The way these tools can be coupled to one another needs careful attention, and will depend on linking model outputs and ecosystem information across a variety of spatial and temporal scales to provide management advice at the level of strategic management planning (Jarre et al. in press [2006]). The choice of models to be used in an “ecoscope” should be driven by the constraints imposed by each model, how its outputs can be linked to (or strengthen) those of other models or tools being considered, and of course to the particular management objectives (and time scales) that have been set.

1034

SHANNON ET AL.

Figure 9. Schematic diagram showing how the models applied in the southern Benguela can be considered along a food-space continuum. Note that single-species models do not quantify trophic or spatial aspects and therefore they are to be considered outside of this continuum but with inputs to and from ecosystem/multispecies models.

Many tools, information systems, and models have been developed particularly during the last decade, such as coastal hydrodynamic models, biogeochemical models (e.g., those developed to explore anoxic events that have implications for the commercially valuable Cape hake species off Namibia [P. Monteiro, CSIR, personal communication]), individualbased models that couple physics and ecology, GIS and various kinds of trophic and ecosystem models. These techniques and models operate at several spatial and temporal scales, and are sometimes highly sophisticated. They address one particular aspect or question, not

the whole ecosystem dynamics. There is no single model or approach in existence that can provide such information. However, together this whole range of models offers a unique opportunity in ecology to address the complexity of marine ecosystems in a diverse and contrasted manner. Despite the variety of techniques that can help us to track spatial and dynamic changes in ecosystems, they are not usually strongly connected to solve scientific questions or to respond to important societal questions. This paper aims to show that we need to and should come at the EAF approach from a multifaceted perspective. With

1035

BENGUELA ECOSYSTEM MODELING

respect to this, many different models may complement one another (or interlink) in achieving specified objectives, in such a manner as to increase our confidence in the information that they are able to provide for fisheries management. For example, using a spatialized, size-based IBM of the southern Benguela ecosystem, Shin et al. (2004) ran comparable altered fishing simulations to those undertaken by Shannon (2001), who used a nonspatial and species-specific (as opposed to size-based) trophic model. Simulation results obtained using these two totally different approaches were generally consistent and prompted Shin et al. (2004) to conclude that this kind of cross-validation between model outputs is a promising means of evaluating the robustness of model outputs. The viability approach can serve as an alternative means of evaluating assumptions, inputs and results of other modeling approaches. By simultaneously considering multiple constraints and viable system states, the viability approach may help in validating or confirming inputs to, and parameter estimates and coefficients that are output from other trophic models such as EwE models, which have been used as a basis for testing constraints defining viable ecosystem states (Mullon et al. 2004). Cross-linking different modeling approaches might assist in providing practical fisheries management advice. For example, modeling has been undertaken to determine minimum viable population sizes of seabird species of conservation concern (Crawford et al. 2001; see previous section). Fisheries management could, for example, aim to ensure sufficient escapement of fish to maintain populations above these lower limits. Useful information can be provided by GIS models that assess the foraging ranges of seabirds with respect to the densities and spatial distributions of pe-

lagic fish as their prey (Drapeau et al. 2004; Pecquerie et al. 2004, Fernández Moreno 2003) As the feeding success of seabirds is likely to be influenced by the density of their prey (e.g., Crawford 2003), the exclusion of fishing from certain areas may enhance food acquisition and thereby the reproductive success of seabirds. However, because sardine and anchovy are highly mobile species, fishing away from breeding seabird colonies may also influence the density of prey within the foraging ranges of breeding seabirds. This could be explored through IBMs. The ecosystem model “toolbox” appears promising with respect to including ecosystem considerations in fisheries management, and should be used as a basis from which to construct an “ecoscope” to assemble ecosystem knowledge from various available models and studies into a useable form, from which to provide advice on ecosystem management. Nonetheless, the successful implementation of an EAF will require that clear and agreed objectives are set. Implementation of EAF is likely to require addressing a wider set of conflicting objectives as the fullscale of conflicts between different stakeholders, some currently unrecognized or overlooked, within an ecosystem is explored. Reconciling these conflicts will be one of the biggest challenges to implementation of EAF and the application of suitable objectivedriven modeling approaches in informing the development of policy and objectives will be critical in meeting that challenge.

Acknowledgments The modeling work reported on in this paper formed part of the IDYLE Research Unit, a joint research program between South Africa and IRD-France, co-funded by the PNEC program (France) and affiliated to SPACCGLOBEC. IDYLE was dedicated to the study

SHANNON ET AL.

1036

and modeling of marine ecosystems. We gratefully acknowledge collaborative work and interactions with the IRD-IDYLE team of researchers, in particular Christian Mullon, Laurent Drapeau and Yunne Shin.

References Aubin, J. P. 1991. Viability theory. Birkhauser and Basel, Boston and Berlin. Aubin, J. P. 1996. Dynamical economic theory: a viability approach. Springer, Heidelberg, Germany. Butterworth, D. S., and R. B. Thomson. 1995. Possible effects of different levels of krill fishing on predators – some initial modeling attempts. CCAMLR Science 2:79–97, Tasmania, Australia. Christensen, V., and D. Pauly. 1992. ECOPATH II—a software for balancing steady-state ecosystem models and calculating network characteristics. Ecological Modeling 61:169–185. Christensen, V., and C. J. Walters. 2004. Ecopath with Ecosim: methods, capabilities and limitations. Ecological Modeling 172:109–139. Cochrane, K. L., D. S. Butterworth, J. A. A. De Oliveira, and B. A. Roel. 1998. Management procedures in a fishery based on highly variable stocks and with conflicting objectives: experiences in the South African pelagic fishery. Reviews in Fish Biology and Fisheries 8:177–214. Cochrane, K. L. 2001. The use of ecosystem models to investigate ecosystem-based management strategies for capture fisheries: introduction. Pages 5–10 in T. Pitcher and K. Cochrane, editors. The use of ecosystem models to investigate multispecies management strategies for capture fisheries. Fisheries Centre Research Reports 10(2):5–10, Vancouver. Crawford, R. J. M. 1998. Responses of African penguins to regime changes of sardine and anchovy in the Benguela system. South African Journal of Marine Science 19:355–364. Crawford, R. J. M. 1999. Seabird responses to long-term changes of prey resources off southern Africa. Pages 688–705 in N. J. Adams and R. H. Slotow, editors. Proceedings of the 22nd International Ornithological Congress, Durban, 1998. BirdLife, Johannesburg, South Africa. Crawford, R. J. M. 2003. Influence of food on numbers breeding, colony size and fidelity to localities of swift terns in South Africa’s western Cape, 1987–2000. Waterbirds 26(1):44–53.

Crawford, R. J. M. 2004. Accounting for food requirements of seabirds in fisheries management—the case of the South African purse-seine fishery.Pages 197– 203 in L. J. Shannon, K. L. Cochrane, and S. C. Pillarand, editors. Ecosystem approaches to fisheries in the southern Benguela. South African Journal of Marine Science 26:197–203. Crawford, R. J. M., P. G. Ryan, and A. J. Williams. 1991. Seabird consumption and production in the Benguela and western Agulhas ecosystems. South African Journal of Marine Science 11:357–375. Crawford, R. J. M., D. M. Allwright, and C. W. Heyl. 1992a. High mortality of Cape Cormorants (Phalacrocorax capensis) off western South Africa in 1991 caused by Pasteurella multocida. Colonial Waterbirds 15(2):236–238. Crawford, R. J. M., L. G. Underhill, C. M. Raubenheimer, B. M. Dyer, and J. Martin. 1992b. Top predators in the Benguela ecosystem—implications of their trophic position. South African Journal of Marine Science 12:675–687. Crawford, R. J. M., J. H. David, L. J. Shannon, J. Kemper, N. T. W. Klages, J. P. Roux, L. G. Underhill, V. L. Ward, A. J. Williams, and A. C. Wolfaardt. 2001. African penguins as predators and prey—coping (or not) with change. South African Journal of Marine Science 23:435–447. Crawford, R. J. M., L. J. Shannon, and P. A. Whittington. 1999. Population dynamics of the African penguin at Robben Island. Marine Ornithology 27:135–143. Cury, P., A. Bakun, R. J. M. Crawford, A. Jarre-Teichmann, R. Quinones, L. J. Shannon, and H. M. Verheye. 2000. Small pelagics in upwelling systems: patterns of interaction and structural changes in “wasp-waist” ecosystems. ICES Journal of Marine Science 57:603– 618. Cury, P. M. 2004. Tuning the ecoscope for the ecosystem approach to fisheries. Pages 272–275 in H. I. Browman and K. I. Stergiou, editors. Perspectives on ecosystem-based approaches to the management of marine resources. Marine Ecoogy Progress Series 274. Cury, P., P. Fréon, C. Moloney, L. Shannon, and Y. Shin. 2005a. Processes and patterns of interactions in marine fish populations: an ecosystem perspective. Pages 475–554 in A. Robinson and K. Brink, editors. The sea: ideas and observations on progress in the study of the seas. The global coastal ocean: multiscale interdisciplinary processes Harvard University Press, Cambridge, Massachusetts. Cury, P., C. Mullon, S. M. Garcia, and L. J. Shannon. 2005b. Viability theory for an ecosystem approach to fisheries. ICES Journal of Marine Science 62:577–584.

BENGUELA ECOSYSTEM MODELING

De Oliveira, J. A. A. 2003. The development and implementation of a joint management procedure for the South African pilchard and anchovy resources. Doctoral dissertation., University of Cape Town, Cape Town, South Africa. De Oliveira, J. A. A., D. S. Butterworth, B. A. Roel, K. L. Cochrane, and J. P. Brown. 1998. The application of a management procedure to regulate the directed and bycatch fishery of South African sardine Sardinops sagax. Pages 449–469 in S. C. Pillar, C.L. Moloney, A. I. L. Payne, and F. A. Shillington, editors. Benguela dynamics: impacts of variability on shelf-sea environments and their living resources. South African Journal of Marine Science 19. Drapeau, L., L. Pecquerie, P. Fréon, and L. J. Shannon. 2004. Quantification and representation of potential spatial interactions in the Southern Benguela ecosystem. Pages 141–159 in L. J. Shannon, K. L. Cochrane, and S. C. Pillar, editors. An ecosystem approach to fisheries in the southern Benguela. African Journal of Marine Science 26. Fernández Moreno, A. 2003. Mapping and quantification of the potential interactions of top-predators in the Southern Benguela Ecosystem. DESS Report, Université du littoral, Calais, France. Fréon, P., L. Drapeau, J. David, A. Fernández Moreno, R. Leslie, H. Oosthuizen, L. J. Shannon, and C. D. van der Lingen. 2005. Spatialised ecosystem indicators in the Southern Benguela. Quantitative ecosystem indicators for fisheries management. ICES Journal of Marine Science 62:459–468. Geromont, H. F., J. A. A. De Oliveira, S. F. Johnston, and C. L. Cunningham. 1999. Development and application of management procedures for fisheries in southern Africa. ICES Journal of Marine Science 56:795– 796. Heath, M. R., and A. Gallego. 1997. From the biology of the individual to the dynamics of the population: bridging the gap in fish early life studies. Journal of Fish Biology 51(Supplement A):1–29. Hinckley, S., A. L. Hermann, K. L. Mier, and B. Megrey. 2001. Importance of spawning location and timing to successful transport to nursery areas: a simulation study of Gulf of Alaska walleye pollock. ICES Journal of Marine Science 58:1042–1052. Huggett, J., P. Fréon, C. Mullon, and P. Penven. 2003. Modelling the transport success of anchovy Engraulis encrasicolus eggs and larvae in the southern Benguela: the effect of spatio-temporal spawning patterns. Marine Ecology Progress Series 250:247–262. Jarre-Teichmann, A., L. J. Shannon, C. L. Moloney, and P. A. Wickens. 1998. Comparing trophic flows in the

1037

southern Benguela to those in other upwelling ecosystems. Pages 391–414 in S. C. Pillar, C. L. Moloney, A. I. L. Payne, and F. A. Shillington, editors. Benguela dynamics: impacts of variability on shelf-sea environments and their living resources. South African Journal of Marine Science 19. Jarre, A., C. L. Moloney, L. J. Shannon, P. Fréon, C. D. van der Lingen, H. Verheye, L. Hutchings, J.- P. Roux, and P. Cury. 2006. Developing a basis for detecting and predicting long-term ecosystem changes. Pages 239-272 in L. V. Shannon, G. Hempel, P. MalanotteRizzoli, C. L. Moloney and J. Woods, editors. Elsevier, Amsterdam. Miller, D. C. M., C. L. Moloney, C. D. van der Lingen, C. Lett, C. Mullon, and J. G. Field. 2006. Modelling the effects of physical-biological interactions and spatial variability in spawning and nursery areas on transport and retention of sardine Sardinops sagax eggs in the southern Benguela ecosystem. Journal of Marine Systems 61(3–4)212–229. Moloney, C. L., C. D. Van der Lingen, L. Hutchings, and J. G. Field. 2004. Contributions of the Benguela ecology programme to pelagic fisheries management in South Africa. Pages 37–51 in L. J. Shannon, K. L. Cochrane, and S. C. Pillar, editors. Ecosystem approaches to fisheries in the southern Benguela. African Journal of Marine Science 26:37–51. Moloney, C. L., A. Jarre, H. Arancibia, Y. M. Bozec, S. Neira, J. P. Roux, and L. J. Shannon. 2005. Comparing the Benguela and Humboldt marine upwelling ecosystems with indicators derived from inter-calibrated models. ICES Journal of Marine Science 62 (3):493–502. Mori, M., and D. S. Butterworth. 2004. Consideration of multi-species interactions in the Antarctic: a preliminary model of the minke whale–blue whale–krill interaction. Pages 245–259 in Shannon, L.J., K. L. Cochrane, and S. C. Pillar, editors. Ecosystem approaches to fisheries in the southern Benguela. African Journal of Marine Science 26:245–259. Mullon, C., P. Cury, and P. Penven. 2002. Evolutionary individual-based model for the recruitment of the anchovy in the southern Benguela. Canadian Journal of Fisheries and Aquatic Sciences 59:910–922. Mullon, C., P. Fréon, C. Parada, C. D. van der Lingen, and J. A. Huggett. 2003. From particles to individuals: modeling the early stages of anchovy in the southern Benguela. Fisheries Oceanography 12(4, 5): 396–406. Mullon, C., P. M. Cury, and L. J. Shannon. 2004. Viability model of trophic interactions in marine ecosystems. Natural Resource Modelling 17(1):27–58. Parada, C., C. D. van der Lingen, C. Mullon, and P. Penven.

1038

SHANNON ET AL.

2003. Modeling the effect of buoyancy on the transport of anchovy (Engraulis capensis) eggs from spawning to nursery grounds in the southern Benguela: an IBM approach, Fisheries Oceanography 12:170– 184. Parada, C. E. 2003. Modeling the effects of environmental and ecological processes on the transport, mortality, growth and distribution of early stages of Cape anchovy (Engraulis encrasicolus) in the Benguela system. Doctoral dissertation. University of Cape Town, Cape Town, South Africa. Pecquerie, L., L. Drapeau, P. Fréon, J. Coetzee, R. Leslie, and M. Griffiths. 2004. Distribution patterns of key fish species of the southern Benguela ecosystem: an approach combining fishery-dependant and fishery-independent data. Pages 115–139 in L. J. Shannon, K. L. Cochrane, and S. C. Pillar, editors. Ecosystem approaches to fisheries in the southern Benguela. African Journal of Marine Science 26:115–139. Penven, P., C. Roy, G. B. Brundritt, A. Colin de Verdière, P. Fréon, A. S. Johnson, J. R. E. Lutjeharms, and F. A. Shillington. 2001. A regional hydrodynamic model of upwelling in the southern Benguela. South African Journal of Science 97:472–475. Punt, A. E., and D. S. Butterworth. 1995. The effect of future consumption by the Cape fur seal in catches and catch rates of the Cape hakes. Modelling the biological interaction between Cape fur seals Arctocephalus pusillus pusillus and the Cape hakes Merluccius capensis and M. paradoxus. South African Journal of Marine Science 16:255–285. Shannon, L. J., 2001. Trophic models of the Benguela upwelling system: towards an ecosystem approach to fisheries management. Doctoral dissertation. University of Cape Town, Cape Town, South Africa. Shannon, L. J., and R. J. M. Crawford. 1999. Management of the African Penguin Spheniscus demersus— insights from modeling. Marine Ornithology 27:119– 128. Shannon, L. J., P. M. Cury, and A. Jarre. 2000. Modelling effects of fishing in the southern Benguela ecosystem. ICES Journal of Marine Science 57:720–722. Shannon, L. J., C. L. Moloney, A. Jarre, and J. G. Field. 2003. Trophic flows in the southern Benguela during the 1980s and 1990s. Journal of Marine Systems 39:83–116. Shannon, L. J. and C. L. Moloney. 2004. An ecosystem framework for fisheries management in the southern Benguela upwelling system. Pages 63–77 in L. J. Shannon, L. J., K. L. Cochrane, and S. C. Pillar, editors. Ecosystem approaches to fisheries in the southern

Benguela. African Journal of Marine Science 26:63– 77. Shannon, L. J., K. L. Cochrane, C. L. Moloney, and P. Fréon. 2004a. Ecosystem approach to fisheries management in the southern Benguela: a workshop overview. Pages 1–8 in L. J. Shannon, K. L. Cochrane, and S. C. Pillar, editors. Ecosystem approaches to fisheries in the southern Benguela. African Journal of Marine Science 26:1–8. Shannon, L. J., K. L. Cochrane, and S. C. Pillar editors. 2004b. Ecosystem approaches to fisheries in the southern Benguela. African Journal of Marine Science 26. Shannon, L. J., V. Christensen, and C. Walters. 2004c. Modelling stock dynamics in the southern Benguela ecosystem for the period 1978–2002. Pages 179–196 in L. J. Shannon, K. L. Cochrane, and S. C. Pillar, editors. Ecosystem approaches to fisheries in the southern Benguela. African Journal of Marine Science 26:179–196. Shin, Y. J., and P. Cury. 2001. Exploring fish community dynamics through size-dependent trophic interactions using a spatialized individual-based model. Aquatic Living Resources 14:65–80. Shin, Y. J., L. J. Shannon, and P. Cury. 2004. Simulations of fishing effects on the southern Benguela fish community using the individual-based model: learning from a comparison with ECOSIM. Pages 95–114 in L. J. Shannon, K. L. Cochrane, and S. C. Pillar, editors. Ecosystem approaches to fisheries in the southern Benguela. African Journal of Marine Science 26:95–114. Starfield, A. M., P. A. Shelton, J. G. Field, R. J. M. Crawford, and M. J. Armstrong. 1988. Note on a modelling schema for renewable resource problems. South African Journal of Marine Science 7:299–303. Ulanowicz, R. E. 1993. Inventing the ecoscope. Pages ix– x in V. Christensen and D. Pauly, editors. Trophic models of aquatic ecosystems. ICLARM Conference Proceedings 26, Tasmania, Australia. van der Lingen, C. D., and J. A. Huggett. 2003. The role of ichthyoplankton surveys in recruitment research and management of South African anchovy and sardine. Pages 303–343 in H. I. Browman and A. B. Skiftesvik, editors. The big fish bang. Procedings of the 26th Annual Larval Fish Conference. IMR, Bergen, Norway. Werner, F. E., R. I. Perry, R. G. Lough, and C. E. Naximie. 1996. Trophodynamic and advective influences on Georges Bank larval cod and haddock. Deep Sea Researh Part II Tropical Studies in Oceanography 43:1793–1822. Walters, C., V. Christensen, and D. Pauly. 1997. Structur-

BENGUELA ECOSYSTEM MODELING

ing dynamic models of exploited ecosystems from trophic mass-balance assessments. Reviews in Fish Biology and Fisheries 7:139–172, Vancouver. Yemane, D. G., J. G. Field, and M. Griffiths. 2004. Effects of fishing on the size and dominance structure of line

1039

fish communities of the Cape Region, South Africa. Pages 161–177 in L. J. Shannon, K. L. Cochrane, and S. C. Pillar, editors. Ecosystem approaches to fisheries in the southern Benguela. African Journal of Marine Science 26:161–177.

Reconciling Fisheries with Conservation: Proceedings of the Fourth World Fisheries Congress

Cover art: “Respect” created for the Fourth World Fisheries Congress by Calvin Hunt. The theme of the Congress is reflected by the title—the reconciliation of fisheries is the expression of “respect.” Embracing the colors of the nations of the world, the woman in the center embodies the spirit of Mother Earth. The eagle on her chest is a crest recognized for its strength and courage. Unity, coming together with one voice, is represented in the circles she holds. The number four, sacred in Northwest Coast native culture, is reflected in the four salmon surrounding the woman, symbolizing four directions, four seasons, and four stages of life.

Reconciling Fisheries with Conservation: Proceedings of the Fourth World Fisheries Congress

Edited by Dr. Jennifer L. Nielsen USGS Alaska Science Center 1011 East Tudor Road Anchorage, Alaska 99503, USA Dr. Julian J. Dodson Department of Biology, University Laval Ste-Foy, Quebec G1K 7P4, Canada Dr. Kevin Friedland National Marine Fisheries Service Blaisdell House, University of Massachusetts Amherst, Massachusetts 01003, USA Dr. Troy R. Hamon National Park Service Post Office Box 7, King Salmon Mall Suite 101 King Salmon, Alaska 99613, USA Dr. Jack Musick Virginia Institute of Marine Science 7036 Sassafras Landing Gloucester, Virginia 23061, USA Dr. Eric Verspoor Freshwater Fisheries Programme FRS Freshwater Laboratory Pitlochry PH16 5LB, Scotland, UK

American Fisheries Society Symposium 49 Proceedings of the Fourth World Fisheries Congress Held in Vancouver, British Columbia, Canada 2 May–6 May 2004 American Fisheries Society Bethesda, Maryland 2008

The American Fisheries Society Symposium Series is a registered serial. Suggested citation formats follow. Entire Book Nielsen, J. L., J. J. Dodson, K. Friedland, T. R. Hamon, J. Musick, and E. Verspoor, editors. 2008. Reconciling fisheries with conservation: proceedings of the Fourth World Fisheries Congress. American Fisheries Society, Symposium 49, Bethesda, Maryland. Chapter within the Book Adlerstein, S., F. Alvarez, and R. Goñi. 2008. Quality data versus analysis? Abundance indices from a Mediterranean trawl fishery as case study. Pages 1107–1120 in J. L. Nielsen, J. J. Dodson, K. Friedland, T. R. Hamon, J. Musick, and E. Verspoor, editors. Reconciling fisheries with conservation: proceedings of the Fourth World Fisheries Congress. American Fisheries Society, Symposium 49, Bethesda, Maryland.

© Copyright 2008 by the American Fisheries Society All rights reserved. Photocopying for internal or personal use, or for the internal or personal use of specific clients, is permitted by the American Fisheries Society (AFS) provided that the appropriate fee is paid directly to Copyright Clearance Center (CCC), 222 Rosewood Drive, Danvers, Massachusetts 01923, USA; phone: 978-750-8400. Request authorization to make multiple copies for classroom use from CCC. These permissions do not extend to electronic distribution or long-term storage of multiple articles or to copying for resale, promotion, advertising, general distribution, or creation of new collective works. For such uses, permission or license must be obtained from AFS. Printed in the United States of America on acid-free paper Library of Congress Control Number 2008920748 ISBN 978-1-888569-80-3 ISSN 0892-2284 American Fisheries Society Web site: http://www.fisheries.org American Fisheries Society 5410 Grosvenor Lane, Suite 110 Bethesda, Maryland 20814-2199 USA