Stock Identification Methods: Applications in Fishery Science

By Stefano Mariani

Stock Identification Methods, 2e, continues to provide a comprehensive review of the various disciplines used to study the population structure of fishery resources. It represents the worldwide experience and perspectives of experts on each method, assembled through a working group of the International Council for the Exploration of the Sea. The book is organized to foster interdisciplinary analyses and conclusions about stock structure, a crucial topic for fishery science and management.

Technological advances have promoted the development of stock identification methods in many directions, resulting in a confusing variety of approaches. Based on central tenets of population biology and management needs, this valuable resource offers a unified framework for understanding stock structure by promoting an understanding of the relative merits and sensitivities of each approach.

• Describes 18 distinct approaches to stock identification grouped into sections on life history traits, environmental signals, genetic analyses, and applied marks
• Features experts' reviews of benchmark case studies, general protocols, and the strengths and weaknesses of each identification method
• Reviews statistical techniques for exploring stock patterns, testing for differences among putative stocks, stock discrimination, and stock composition analysis
• Focuses on the challenges of interpreting data and managing mixed-stock fisheries

• Language: English
• Publisher: Academic Press
• Release date: Oct 4, 2013
• ISBN: 9780123972583

Book preview

Sinclair

Introduction

Stock identification involves delineation of population structure of fishery resources—a central theme in fisheries science and management. Developments in molecular biology and chemistry and advances in image analysis and tagging technologies have prompted revolutionary changes in many stock identification approaches. Stock identification is developing as an increasingly interdisciplinary field and has become a requisite component of fishery science and management programs performed worldwide by research institutions and government agencies.

Continued sophistication of technological aspects across disciplines (from genomics to modeling) and improved understanding of biological and environmental processes in the oceans have led to a dramatically reappraised second edition of Stock Identification Methods. This book provides a new outlook on recently developed techniques, and continues to provide guidance on how to integrate information from multiple stock identification approaches and draw holistic and robust conclusions that have practical implications for fisheries management and conservation biology.

Authored by experts on the topic who have experience in the application of stock identification techniques, this book is designed to serve as a valuable guide to individual researchers and managers as well as graduate students.

New to this Edition

• New synthesis on the use of latest-generation nuclear genetic markers

• Presents new evaluation of approaches for the integration of multidisciplinary data

• Highlights the rapid advancements in electronic tagging technology and technological developments in acoustic and data storage tags

• Features new content on use of models to synthesize information gained from multiple stock identification methods

Chapter One

Stock Identification Methods

An Overview

Steven X. Cadrin¹, Lisa A. Karr¹,² and Stefano Mariani³,    ¹University of Massachusetts, School for Marine Science and Technology, New Bedford, MA, USA,    ²Gulf of Maine Research Institute, Portland, ME, USA,    ³School of Environment and Life Sciences, University of Salford, Manchester, United Kingdom

Abstract

Stock identification involves delineation of population structure of fishery resources—a central theme in fisheries science and management. Developments in molecular biology and chemistry and advances in image analysis and tagging technologies have prompted revolutionary changes in many stock identification approaches. Stock identification is developing as an increasingly interdisciplinary field and has become a requisite component of fishery science and management programs performed worldwide by research institutions and government agencies.

Continued sophistication of technological aspects across disciplines (from genomics to modeling) and improved understanding of biological and environmental processes in the oceans have led to a dramatically reappraised second edition of Stock Identification Methods. This book provides a new outlook on recently developed techniques, and continues to provide guidance on how to integrate information from multiple stock identification approaches and draw holistic and robust conclusions that have practical implications for fisheries management and conservation biology.

Keywords

Glossary; Methods; Stock Identification

Chapter Outline

Glossary

Acknowledgments

References

Acknowledgments

We thank the chapter authors for their dedication of time, energy, and expertise to this initiative. We also thank Kevin Friedland and John Waldman for their leadership in producing the first edition. The ICES Stock Identification Working Group and other collaborators have been an active influence on our thoughts and the organization of this book. We thank ICES for their continued support and the staff at Elsevier for their help.

Stock identification is a central theme in fisheries science that involves the recognition of self-sustaining components within natural populations. Stock identification is a prerequisite for the tasks of stock assessment and fishery management because most applied population models assume that the group of individuals has homogeneous vital rates (e.g., growth, maturity, mortality) and a closed life cycle, in which young fish in the group were produced by previous generations within the same group.

Secondary roles for stock identification in fishery science are also important but less obvious. Any study that wishes to represent a living resource through field sampling, or even laboratory studies, should consider the species' population structure in the sampling and analytical design. Whether the research concerns general life history, growth, physiology, or diet, the population of inference and its stock components should be identified. Therefore, stock identification can be viewed as a prerequisite for any fishery analysis, just as population structure is considered a basic element of conservation biology (Crandall et al., 2000; Thorpe et al., 1995).

Despite its importance in the development of fishery advice and management, stock identification continues to be an afterthought. The fishery science community has a habit of building assessments from back to front, often only giving cursory treatment to stock identification, and in the name of being expeditious, population vital rates are estimated without regard to lingering questions about stock structure. We hope that this volume will not only provide source material to improve the quality of stock identification research but also may stimulate new research on stocks being assessed without the benefit of reliable stock identification.

One reason for the reluctance to prioritize stock identification is that it remains one of the most confusing subjects in fisheries science, with a wide variety of approaches, rapidly advancing methodologies, challenges in sampling, as well as conflicting terminologies and interpretations. There have been some excellent reviews of stock identification research, including concise overviews (Simon and Larkin, 1972; Templeman, 1982; Pawson and Jennings, 1996; Waldman, 1999) and conference proceedings that include various case studies (Ihssen et al., 1981; Kumpf et al., 1987; Begg et al., 1999). However, many of the case studies on stock identification are result oriented and narrowly focused, and overview perspectives lack the detail needed to guide researchers.

In 1992 the International Council for the Exploration of the Sea (ICES) established a Study Group on Stock Identification Protocols for Finfish and Shellfish Stocks to review methodologies of stock identification and develop a protocol for the application of stock identification results. The support of this work by the ICES community reflects the continuing leadership of ICES in oceanographic and fishery research. Coincidentally, it was an ICES committee that first promoted stock identification as an important consideration for fishery science in the early 1900s (Smith, 1994). The ICES Study Group was organized in an open format to invite a wide participation of experts on stock identification to summarize the various approaches. Over the following decade, the group expanded to the Study Group on Stock Identification Methodology and developed a volume of contributions to review each existing method, with emphasis on recent advances, review of benchmark case studies, critique of strengths and weaknesses, and guidance for effective protocols. The compilation provided the basis for the first edition of this book (Cadrin et al., 2005). The compendium of methodological reviews was designed to serve as a resource for researchers interested in comparative studies in stock identification as well as a general introduction for all scientists and managers of natural resources. Methodological chapters were not necessarily comprehensive reviews but focused more on historical development, benchmark case studies, critique of current issues, and prescriptions for the most effective protocols for stock identification.

One theme that emerged throughout the development of this volume was the strength of interdisciplinary analyses. Over the history of stock identification, new methods were developed and promoted as better ways to approach stock identification, often leading to equivocal information from competing methodological camps. However, when results from each approach are viewed in the context of what precise aspect of stock structure they reveal (defined in this volume), a more holistic view with multiple perspectives is possible, providing more reliable information for resource management. As new methods continue to emerge, their results should be considered along with those from traditional approaches to improve our ability to study stock structure.

After publication of the first edition, the Working Group on Stock Identification Methods continued to serve ICES. Each year the working group summarized advances in the field and formed recommendations on specific stock identification issues related to ICES advice. The second edition was developed to update the information on advancing technologies and interdisciplinary approaches to fish stock identification. Developments in molecular biology, electronic tags, chemical methods, and image analysis have prompted revolutionary changes in many stock identification approaches. In the foreword of the 2005 edition, Mike Sissenwine stated that "Although this volume will be a valuable reference for years to come, I think that we should all be excited by the prospect of innovative advances in the near future that surely will render some of the conclusions in the book out of date. The scientists responsible for the volume, and ICES as the sponsor of the Working Group, do not want to rest on their laurels. Advances in biochemistry, analytical chemistry, and electronics (including microtechnology with nanotechnology on the horizon) foretell a very productive era unfolding when it comes to stock information in support of better science and better resource management. Speaking as both the President of ICES and the Chief Science Advisor for the US National Marine Fisheries Service, I look forward to important and exciting discoveries in the future."

The general goals of the second edition were to: (1) update each chapter with information on the technological and methodological advances in the last decade and (2) address a shortfall of the first edition by providing guidance on how to integrate information from multiple stock identification approaches to draw interdisciplinary conclusions. More specifically, several new genetic techniques have been developed and refined, while others have been so rapidly superseded that they no longer find space in the stock identification methods tool kit. Electronic tagging technology has also rapidly advanced in the last decade with technological developments in acoustic, archival, and satellite tags. Additionally, simulation studies have emerged as a useful tool for integrative analyses and understanding the practical implications of stock structure for fishery management and conservation biology. The revised final section of the book describes applications of stock identification for fishery management and conservation biology through interdisciplinary analysis and synthesis.

This book provides guidance on best practices for stock identification so scientists and managers can confront the complex issues and management challenges that they are faced with. This new edition also offers a glossary at the end of this chapter to standardize terminology and promote consistent use of terms. We hope that the ideas in this volume can be developed and applied in a wide range of scenarios so management units can be practical reflections of biological population structure.

Glossary

Adaptive genetic markers Genetic markers whose variation depends on the action of natural selection.

Biological population A self-sustaining group of individuals, from a single species, whose dynamics are primarily determined by birth and death processes.

Connectivity A link between two life-history stages, biological units, or habitat patches. In this sense the two points can be considered as linked by a thread of varying robustness, which can be traced from one end to the other.

Contingent A group of fish that co-occur in space and time and adhere to the same life history pattern.

Deme A local randomly mating genetic subunit within a species or metapopulation.

Hardy–Weinberg equilibrium (HWE) The condition under which the allele and genotype frequencies in a population will remain constant, assuming large random-mating populations that experience no migration, no mutation, and no natural selection.

Isolation by distance A mechanism of spatial structure according to which individuals are more likely to mate with individuals from nearby populations rather than from distant ones. This model is expected to result in a positive correlation between geographic distance and degree of genetic divergence.

Management unit A geographically delineated fishery resource that is based on practical or jurisdictional boundaries for operational stock assessment and fishery management, which may or may not reflect biological population structure.

Metapopulation A system of interacting biological populations, termed subpopulations, that exhibit a degree of independence in local population dynamics as well as connectivity between subpopulations.

Natal homing A return migration of sexually mature individuals to spawn upon the grounds where they were spawned.

Natural selection The nonrandom process by which phenotypic frequencies change in a population as a result of heritable variation in their fitness.

Neutral genetic markers Genetic markers whose variation is independent of the action of natural selection.

Panmixia The status of a population within which mating is completely random and all adult individuals are potential partners for one another. In population genetic terms, a panmictic population is in Hardy–Weinberg equilibrium.

Reaction norm The pattern of phenotypic expression of a genotype across a range of environmental conditions.

Reproductive mixing (straying, entrainment) Individuals that were spawned in one location (or season) and subsequently spawn in a different location (or in a different season), hence joining a different population or forming a new one. The processes by which they join a different population can be by straying (i.e., random or density-dependent movement to a new location) or being entrained (i.e., following the movement of individuals from a different spawning group during periods of spatial overlap).

Spatial/temporal overlap Co-occurrence of two or more biological units in space and time. Spatial overlap can occur without temporal overlap. For instance, this can happen when spawning grounds are used at different times of the year (e.g., autumn, winter, and/or spring spawning populations).

Spawning component (group, aggregation) A group of individuals that utilizes a single spawning ground. A biological population may comprise a single or a number of different spawning components.

Stock An exploited fishery unit. A stock may be a single spawning component, a biological population, a metapopulation, or comprise portions of these units. For management purposes stocks are considered discrete units, and each stock can be exploited independently or catches can be assigned to the stock of origin.

Subpopulations (components of a metapopulation) A single, mostly self-sustaining unit within a metapopulation.

References

1. Begg G, Friedland KD, Pearce JB. Stock identification – its role in stock assessment and fisheries management. Fish Res. 1999;43:1–8.

2. Cadrin SX, Friedland KD, Waldman J, eds. Stock Identification Methods: Applications in Fishery Science. Elsevier Academic Press 2005.

3. Crandall KA, Bininda-Emonds ORP, Mace GM, Wayne RK. Considering evolutionary processes in conservation biology. Trends Ecol Evol. 2000;15(7):290–295.

4. Ihssen PE, Bodre HF, Casselman JM, McGlade JM, Payne NR, Utter FM. Stock identification: materials and methods. Can J Fish Aquat Sci. 1981;38:1838–1855.

5. Kumpf, H.E., Vaught, R.N., Grimes, C.B., Johnston, A.G., Nakamura, E.L., 1987. Proceedings of the Stock Identification Workshop. NOAA Tech. Mem. NMFS-SEFC-199.

6. Pawson MG, Jennings S. A critique of methods for stock identification in marine capture fisheries. Fish Res. 1996;25:203–217.

7. Simon RC, Larkin PA, eds. The Stock Concept in Pacific Salmon H.R MacMillan Lectures in Fisheries. Univ. British Columbia 1972.

8. Smith TD. Scaling Fisheries: A Science Driven by Economics and Politics 1855–1955. Cambridge University Press 1994.

9. Templeman W. Stock Discrimination in Marine Fishes. 1982; NAFO SCR Doc. 82/IX/79.

10. Thorpe J, Gall G, Lannan J, Nash C. Conservation of Fish and Shellfish Resources: Managing Diversity. Academic Press 1995.

11. Waldman JR. The importance of comparative studies in stock analysis. Fish Res. 1999;43:237–246.

Chapter Two

The Unit Stock Concept

Bounded Fish and Fisheries

David H. Secor,    Chesapeake Biological Laboratory, University of Maryland Center for Environmental Science, Solomons, Maryland, USA

Abstract

The geographic ambit of human influences in the marine environment has increased over the hundreds of years that fishers and scientists have observed the linked cycles of fish migration and abundance. Codified as the unit stock, modern fisheries science has been motivated to define groups of fish that are subject to geographically bounded influences: fishing, hatchery augmentation, pollution, habitat loss, and climate. Although stocks are often defined as reproductively isolated populations, in many instances group entities (e.g., contingents and subpopulations) other than populations are most influenced. Further, structural elements within populations often contribute to yield, stability, persistence, and resilience through connectivity and aggregate metapopulation responses. Flexible (operational) definitions of stocks retain important ecological meaning, accommodate the complex life cycles of marine fishes, and are required for ongoing challenges related to overfishing and recovery of overfished stocks, mixed stocks, and interjurisdictional fisheries, climate-shifted stocks, and intensive spatial management.

Keywords

Connectivity; Contingent; Life cycle; Migration; Mixed stock; Natal homing; Population; Practical management unit; Shifted stock; Unit stock

Chapter Outline

2.1 The Unit Stock Imperative

2.2 Operational Definitions of Unit Stock

2.3 Fishing across Boundaries

2.4 Mixed and Shifting Stocks

2.5 Complex Life Cycles

2.6 Stocks as Closed Populations

2.7 Natal Homing Mechanisms

2.8 Self-Recruitment in Reef Fishes

2.9 Open Populations

2.10 Between Closed and Open Populations: Connectivity

2.11 What Do We Need to Know to Track Fish Stocks?

References

Further Reading

2.1 The Unit Stock Imperative

For hundreds of years, the seasonal and annual cycles of capture fisheries have prompted observation, speculation, and theories on fish migration. Early communication networks allowed the seasonal displacement of species such as Atlantic bluefin tuna, Atlantic cod, and Atlantic and Pacific salmons to be followed as they migrated across traditional fishing grounds (Aristotle, 350 B.C.; Hjort, 1914; Gilbert, 1915; Ravier and Fromentin, 2004). In the early nineteenth century, the philosopher Chong (1814) noted cycles of herring abundance in two regions along the Korean Peninsula that were offset from each other. Further, herring from either region exhibited substantial differences in vertebral counts. This led Chong to conceive that migrations were limited so that groups of herring were spatially discrete, each exhibiting a unique cycle of abundance. Later, Heincke (1898) quantified regional morphometric differences in Atlantic herring in the North Sea, which he ascribed to curtailed migration by local populations (races in the late nineteenth-century lexicon). This view supplanted earlier ideas of a single large panmictic group of Atlantic herring interacting with fisheries throughout the North Sea. Similarly, Gilbert (1915) in British Columbia, Canada, used fish scale morphometrics of sockeye salmon to give evidence of separate race dynamics specific to each spawning tributary. Heincke, Gilbert, and other early observers of local population structure argued that the collective movements of individuals, organized as races, should make them differentially vulnerable to fishing and environmental influences. For instance, Heincke believed that the siting of a whale rendering plant had specific effects on a single proximate race of herring (Smith, 1994). Gilbert described how collapse of canyon walls into Hell's River Gorge, due to overzealous use of explosives in laying train rails, fully impeded the spawning runs and led to the catastrophic loss of a race of Fraser River sockeye salmon. In Norway, early hatchery proponents argued that fisheries on local races of fjord cod could be efficiently augmented through release of hatchery-produced larvae (Solemdal et al., 1984).

Building on this early concept of local influence on local races, modern fisheries science has been largely motivated by the need to understand how fishing, climate, habitat loss, pollution, and hatchery augmentation—forces that are bounded geographically—influence the internal dynamics of certain groups of individuals. This required two central inquiries: (1) which group of individuals is subject to these forces, and (2) having identified the affected group, how are its internal dynamics affected? In an early quantitative treatment on geographic scales of influence, Dahl (1909) refuted the premise that cod hatcheries effectively augmented catches in individual Norwegian fjords. His group did so by showing that the internal dynamics (measured by juvenile abundances in fjords) were not related to how many cod were stocked into each fjord. Rather, cod showed correlated dynamics across fjords, which indicated broader regional-scale population dynamics (Secor, 2005). In another classic example, Hjort and his colleagues (Hjort and Lea, 1914) discovered the dominant year-class phenomenon based on the age structure of Atlantic herring (see Sinclair, Forward). A critical observation was that the same dominant year-class was apparent regardless of where herring schools were sampled up and down the Norwegian coast. This suggested that climate and fishing influences were integrated by a larger group of fish, one which ranged across the entire Norwegian Sea. Examination of age structure within and between groups of fish remains a critical measure of group cohesion. For instance, the apparent disappearance and recovery of winter skate on George's Bank during the period 1980–2000, thought to be due to population abundance dynamics, was later shown by investigating age structure to be due to a transient northern shift in their distribution. Frisk et al. (2008) discovered that during the recovery period skates suddenly emerged with a fully developed age structure similar to the one prior to collapse. They then provided strong circumstantial evidence for a short-term spatial excursion by the same group of skates rather than a change in their internal dynamics.

The stock concept arose due to the need to define discrete groups of fish so that their internal dynamics could be audited against the effects of fishing (Cadrin and Secor, 2009). Thus, central theories of dynamics related to sustainable fisheries—the stock-recruitment and surplus-yield constructs (Smith, 1994)—were attached to certain groups of fish termed stocks. Once these groups were practically defined by when and where they were fished, demographic models such as Russell's catch equation could be applied to audit fishing rates against rates of natural renewal. Such groups were termed unit stocks, emphasizing their discrete internal responses to fishing and other outside forces, such as climate. Early definitions on the unit stock were decidedly agnostic about why groups might exhibit distinct internal dynamics and emphasized practical definitions based on fishery characteristics (aka the harvest stock: Russell, 1931; Cope and Punt, 2009) and homogeneity of demographic attributes (Waples et al., 2008). The practicality of managing fisheries has caused these definitions to persist. For instance, FAO defines stock as a sub-set of one species having the same growth and mortality parameters, and inhabiting a particular geographic area (FAO, 1998). Still, because stock-recruitment and surplus-production models both specify theories of population replacement, it was inevitable that unit stock and biological population (discrete self-reproducing group; sensu, Marr, 1957) would be thought of as one and the same (Waldman, 2005). Thus, most efforts to define stocks over the past 100 years have depended on methodological approaches designed to evaluate reproductive isolation (e.g., tagging, genetics). One need only review the concepts and approaches detailed in this volume to appreciate the prevalence of population thinking in the concept of stock.

2.2 Operational Definitions of Unit Stock

Here, as previously (Secor, 2005), it is argued that stocks (i.e., groups that exhibit unique demographic dynamics to external forcing) cannot be conceived exclusively as biological populations. The reason is twofold: (1) fishing and other external influences occur at multiple scales such that groups other than populations are often the most affected; and (2) the internal dynamics within populations and other groups are not homogenous nor are they exclusively internal. These justifications are supported by decades of discovery on the individual movements and collective migrations of marine fishes driven by developments in electronic tagging, life cycle tracers (aka natural tags), oceanographic observing systems, and modeling (Secor, in preparation). New understandings on the diversity of life cycles that underlie population structure argue against a strict typological view, where the internal dynamics of each unit stock can be simply parameterized and modeled (Cadrin and Secor, 2009; Petitgas et al., 2010; MacCall, 2012). Further, fisheries science and management has broadened considerably from traditional goals based on yield to those that incorporate stability and resilience, ecosystem considerations, and fishery and management feedbacks (adaptive management). In this broader framework, flexible models can incorporate multiple scales and sources of information to address how diversity in spatial behaviors affects fishery population dynamics (see Chapter 21, Kerr and Goethel).

If we define stocks by the geographic extent of fishing or other influences of interest, then stock is being operationally defined—that is, by the question we are asking or by other practical considerations such as the method employed. Booke (1981) initially proposed this flexible construct, where stock was a species group, or population of fish that maintains and sustains itself over time in a definable area. Importantly, an operational definition need not be arbitrary in terms of population or species biology. Rather, the question at hand decides the level of ecological entity, which can range from species to brood (Figure 2.1). For instance, conservation biology typically focuses on the persistence and recovery of species, metapopulations, and populations. The Distinct Population Segment of the U.S. Endangered Species Act spans these levels of organization but also brings in practical elements such as governmental boundaries (USFWS, 1996; Ford, 2004). Similarly, the U.S. Magnuson-Stevens Fisheries Conservation Act defines stock as a species, subspecies, geographical grouping, or other category of fish capable of management as a unit (NOAA, 2007). Operationally, then, a stock can be defined variously by its ecological, technical, recreational, economic, or fishery attributes. Indeed the current emphasis in stock identification is to undertake a holistic approach that considers complementary approaches that convey information across multiple spatial and temporal scales and/or utilizes redundancy (weight of evidence) in stock identification parameters (Dizon et al., 1992; Begg and Waldman, 1999; Swain et al., 2005).

FIGURE 2.1 Levels of ecological organization relevant to the unit stock. Diagram concept adapted from Secor (2005).

2.3 Fishing across Boundaries

Governments have instituted policy frameworks to prevent overfishing of stocks, including stipulation of thresholds related to overfishing rate and overfished abundance level, and harvest target levels that relate not only to maximum sustainable yield but also to uncertainty and risk of future stock depletion. Further, in overfished stocks, policies require fishing controls that enable timely recovery of biomass. As a result, the health of many fished stocks has improved (Worm et al., 2009; Hutchings et al., 2010; NOAA, 2012). However, in numerous instances, overfished stocks have failed to recover, a principal cause of which has been attributed to mis-specified stock structure (Hutchings et al., 2010; Murawski, 2010; Petitgas et al., 2010). Stock structure, the basis for effective fisheries assessment and management, depends on three criteria: (1) identification of the stock, a group of fish with homogeneous internal dynamics and limited exchange with other stocks; (2) evaluation of the stock unit area, the geographic boundaries associated with the seasonal movements, and concentrations (habitats) of that stock; and (3) long-term stability of the stock and its boundaries (Begg et al., 1999; Cope and Punt, 2009; Link et al., 2011).

How humans encounter fish substantially differs from the way in which groups of fish spatially segregate and interact. Historically, a principal way in which humans interacted with fishes was through systems of customary tenure rights. In coastal villages of Japan, hereditary rights were conferred to individual families to fish in certain coastal regions (Kada, 1984; Kalland 1984). This traditional means of prosecuting fisheries has dramatically changed during the past 150 years as fisheries have become increasingly industrialized, leading to increased vessel travel, more efficient searches, and improved capture techniques. Expanded fishing ranges led to conflict and the need for spatial controls. In nearshore waters, individual and village tenure rights gave way to systems of prefectural control in Japan. Layered systems of coastal zone management now occur globally. As one example in U.S. Atlantic Ocean waters, nearshore state fisheries (three miles from shore) are regulated through the Atlantic States Marine Fisheries Commission, and those occurring 3–200 miles from shore are regulated by U.S. Regional Management Councils. Beyond the 200-mile Economic Exclusive Zone (EEZ), fisheries are regulated according to the United Nations Convention on the Law of the Sea through Regional Fishery Management Organizations such as the Northwest Atlantic Fisheries Organization (NAFO), North East Atlantic Fisheries Commission (NEAFMC), and the International Commission for the Conservation of Atlantic Tunas (ICCAT). Thus zonal management scales from local harbors to global international compacts. Assessing where fish are caught across scales of management is complex and expensive, leading to unit stock boundaries that are often statistical in nature—areas that can be consistently identified and audited (Begg et al., 1999; Cope and Punt, 2009).

Fishery management boundaries, whether termed practical management units, statistical areas, or unit stock areas, are often mismatched with the spatial ambits of populations and other groups of fish. As a recent example, a fishery for the pelagic (or beaked) redfish Sebastes mentella rapidly emerged during the period 1981–1995 (peak harvest 180,322 metric tons), initially developed by the former USSR (subsequently Russia), and joined later by the Faroe Islands, Germany, Greenland, Iceland, and Norway (Sigurðsson et al., 2006). The fishery was centered in the Irminger Sea, a region rife with management boundaries, including those of NAFO, NEAFMC, and the EEZ territorial waters of Greenland and Iceland (Figure 2.2). Two distinct fisheries emerged: one centered in shallower depths <500 m in the central part of the sea and another targeting greater depths in the northwestern Irminger Sea and Icelandic territorial waters. The question of whether these two fisheries represent two populations with discrete depth behaviors has become controversial. A holistic stock structure analysis has recently replaced the initial view of a single stock with a construct of two genetically discrete stocks exhibiting distinct depth-specific reproductive and trophic behaviors (Cadrin et al., 2010, Cadrin et al., 2011; Makhrov et al., 2011). A new management unit for the deep pelagic stock straddles four NAFO, NEAFMC, and EEZ regions and statistical areas (Figure 2.2). Controversy ensued because more stringent fishing controls are required to sustain the two separate stocks of pelagic rockfish than a single contiguous population (Cadrin et al., 2011).

FIGURE 2.2 New management unit (stock unit area; shown as lightly shaded region with fish icon) for pelagic rockfish Sebastes mentella in the Irminger Sea as proposed by Cadrin et al. (2011). The management unit was proposed to accommodate a newly recognized deep pelagic stock (>500 m). Note that the new management unit overlaps four statistical and jurisdictional boundary units. Statistical areas designations are those of the North East Atlantic Fisheries Commission; Economic Exclusive Zones (EEZ) for Greenland, Iceland, and Faroe Islands (to southeast) are shown. Diagram adapted from Cadrin et al. (2011).

2.4 Mixed and Shifting Stocks

Mixed stocks occur when stock separation is incomplete and groups overlap in their spatial range for periods of time. Where fisheries occur on mixed stocks, sustainability of those fisheries depends on the internal dynamics of each stock, weighted by their degree of overlap. Because stock demographics and levels of mixing are difficult to simultaneously assess, they are often ignored, resulting in false apparent trends in fishery assessments. Consider the mixed stock issue for Atlantic bluefin tuna. Mixed-stock fisheries for Atlantic bluefin tuna are documented for the U.S. shelf waters where two populations—originating either in the Gulf of Mexico or Mediterranean—intermingle (Rooker et al., 2008; Secor et al., 2012). The assessment model for the western stock (virtual population analysis) is heavily influenced by U.S. catch indices, which in recent years have shown a strong 2003 year-class (ICCAT, 2011). This is important because recruitment has been persistently low during the past 30 years for the Gulf of Mexico population. But, to determine whether this population has returned to historical levels of high recruitment, the 2003 year-class must be weighted (stock identified) according to its population of origin (Figure 2.3). Preliminary evidence indicates that this year-class is indeed dominated by juveniles produced from the Gulf of Mexico population (Secor et al., 2012). The degree of stock mixing by bluefin tuna in eastern Atlantic waters (European and other Mediterranean nations) remains poorly known, but because the Mediterranean population is approximately tenfold more productive and supports higher fishing rates, catches will disproportionately affect the Gulf of Mexico population, which has been in a long-term but unsuccessful rebuilding plan (Taylor et al., 2011). This is a common attribute of mixed-stock fisheries, where the less productive stock receives higher proportional removals; in these instances, management controls should guard against overharvest of the subordinate stock (Ricker, 1958; Nehlson et al., 1991).

FIGURE 2.3 Diagrams illustrating the interplay between shifts in stock structure and resulting trends in actual and apparent stock abundances. The top panel shows shifting stock dynamics where two populations (P1 and P2) are outlined in solid or dashed lines. These stocks occur in two stock unit areas represented by adjacent boxes A and B. The middle panels show actual stock trends in abundance for the two stocks, and the bottom panel shows apparent abundance dynamics occurring within the two stock unit areas. Specific examples of shifting stocks scenarios—Asymmetric Production, Transient Shift, Contraction, and Irruption—are reviewed in text. Diagram concept is adapted from Link et al. (2011).

Distributional shifts in groups and populations of fishes are frequently noted, caused by a range of factors including changes in climate and ocean states, irruptions (rapid population expansions), overfishing, colonization, and food web changes. Link et al. (2011) provided a set of likely scenarios of how shifts in stock distribution can influence apparent trends in their abundance. Shifted stocks are considered respective to a fixed stock unit boundary (Figure 2.3). Transient shifts can occur due to environmental forcing. As an example, climate warming causes a north temperate stock to shift its distribution northerly across a boundary. In this instance, the original unit stock area will show stock extirpation, whereas the adjacent area—if already occupied by the same species—will show increased abundance. The shifted distribution of Georges Bank winter skate (Frisk et al., 2008) is an example of this phenomenon. Contraction of a stock that originally straddled a management boundary to one or the other side of that boundary can manifest a similar result. An example here is contraction of the shelf-distributed Northern stock of Atlantic cod to small coastal stocks (Rose et al., 2000; Petitgas et al., 2010). If the shifted distribution causes species expansion to new areas, then developing fisheries can emerge as has been the case of albacore and bluefin tuna in the Northeast Atlantic Ocean (Dufour et al., 2010). Depending on the type of shifted pattern, assessments can result in alternatively overly optimistic (e.g., high abundance for a contracted coastal cod stock; novel stocks of albacore tuna) or pessimistic (e.g., vacated Georges Bank winter skate) portrayals of stock abundance (Link et al., 2011).

2.5 Complex Life Cycles

The internal dynamics of stocks are increasingly recognized as non-homogeneous (e.g., multiple maturation schedules within a population) and influenced by migrations that are not consistently bounded by geography (e.g., decadal shifts in distribution). Developments in telemetry and life cycle tracer approaches have provided unprecedented details on individual movement behaviors (Secor, in review). The diverse movements of individuals within populations and other groups contribute to heterogeneous and incompletely bounded internal dynamics of stocks. The recent subdiscipline of movement ecology (Nathan et al., 2008) focuses on what motivates these individual movements but leaves alone the question of how individual movements combine at collective scales and drive internal dynamics of stock entities. Rather, two subdisciplines of population ecology have emerged, relevant to the issue of internal dynamics of stocks: population connectivity (Cowen and Sponaugle, 2009) and migration ecology (Secor, in review). These disciplines are similarly defined by: (1) the collective movements of groups of fish and (2) what happens along the way—internal demographic dynamics. Population connectivity principally considers biphasic life cycles, where population closure depends on larval dispersal (see Self-Recruitment, following). Migration ecology more broadly considers complex life cycles in which the migrations across life stages contribute to the dynamics of populations and other groups. A central question for both these disciplines is whether groups defined by life cycles are completely or incompletely closed to other groups (i.e., closed and open populations).

2.6 Stocks as Closed Populations

Although early investigators recognized the importance of open life cycles (e.g., Hjort, 1914; Cushing, 1962; Harden Jones, 1968), development of fisheries assessment science required assumptions of closed populations leading to a historical emphasis on population thinking in marine fisheries science (Smith, 1994; Cadrin and Secor, 2009). During the recent decades, scientific and management emphasis has been on open rather than closed populations, with conceptual developments focused on metapopulations and population connectivity (see Kritzer and Liu, Chapter 3, this volume). Still, it is important to recognize that many recent discoveries continue to point to the remarkable abilities of marine fishes to find their way back home. Several examples follow.

The constancy in the timing and location of spawning runs has promoted salmon as the model species for concepts related to life cycle closure through natal homing (Quinn, 2005). Fraser River sockeye salmon are among the best-studied species (Gilbert, 1915; Burgner, 1991). In the timing of spawning runs, specificity of spawning sites, and migration of juvenile parr to specific rearing habitats, sockeye salmon epitomize life cycle closure centered on local adaptation. Demersal spawning habitats—streambed or lake beach—can be mere hundreds of meters apart, yet spawners utilizing either habitat are unique in their time of arrival, their body morphology, their mating system, and the emergence time and subsequent migration behaviors of their young. Further, they are genetically discrete despite the potential to (1) mate with spawners in adjacent streams or lakes or (2) migrate and spawn elsewhere in the Fraser River watershed. That a fish spends most of its life migrating thousands of kilometers throughout the Alaska Gyre and then returns and spawns at a fixed site on a fixed itinerary is remarkable. It exemplifies how life cycle closure can cause populations to be tightly coupled to specific environments. Stewart et al. (2003) concluded that in this way, Homing creates reproductive isolation and allows for local adaptation in life history traits.

In their use of very specific sets of demersal spawning sites over specific ranges of dates, Atlantic herring might be imagined as a fully marine fish version of Pacific salmon. Unlike the more robust salmon fry, however, the herring larva that emerges from coastal spawning beds has very little control over its dispersal fate. Still, a large set of spawning habitats are consistently occupied across generations (Iles and Sinclair, 1982; Stephenson et al., 2009). How natal homing operates remains unknown, but a well-demonstrated example of natal homing occurs for Celtic Sea herring, which spawn on shelf banks along the southern coast of Ireland. Brophy et al. (2006) used otolith microstructural analysis (see also, Brophy, Chapter 8) to show that despite larval advection into the neighboring Irish Sea, these progeny consistently returned to the Celtic Sea to spawn as adults. In an early summary of herring population studies, Harden Jones (1968) generalized this pattern of larval drift and adult return as the migration triangle (see also, Secor, 2005). Natal homing by Celtic Sea herring conforms well to the Harden Jones concept of life cycle closure.

Like the Celtic Sea herring, bluefin tuna larvae originating in the Gulf of Mexico are quickly transported away from where they are spawned, limiting the opportunity of larvae to imprint to local oceanographic conditions. Recent life cycle tracer (Rooker et al., 2008; Dickhut et al., 2009) and telemetry (Block et al., 2005; Taylor et al., 2011) approaches have shown that despite the fact that juveniles will frequently undertake trans-Atlantic migrations, spawning adults exhibit natal homing and spawning fidelity, thus exhibiting life cycle closure.

2.7 Natal Homing Mechanisms

Imprinting is the most commonly invoked mechanism controlling natal homing in marine fishes (Cury, 1994). During an early genetically controlled period, larvae or juveniles sense and memorize environmental stimuli, distinctive for their site of natal origin (Stabell, 1984). Imprinting then provides a mechanism for natal homing and subsequent spawning site fidelity (for iteroparous fishes), which over generations leads to philopatry and population structure due to reproductive isolation. For Atlantic herring and Atlantic bluefin tuna, imprinting is not as easily understood as in salmonids, but if we accept such a mechanism, then natal homing should conserve ontogenetic and seasonal migration patterns over generations. In the member-vagrant hypothesis, Sinclair (1988) gave priority to natal homing as a means for populations to exploit oceanographic features that would favor the retention of larvae across generations. Sinclair termed this life cycle selection. Similar to what has been described previously for salmon, the theory posits that spawning at a particular time and place preserves a legacy by individual populations to occupy a specific set of habitats (Figure 2.4).

FIGURE 2.4 Diagrams illustrating two central explanations for closed populations in marine fishes. Top panel represents larval retention, imprinting, and natal homing; bottom panel represents social transmission of adult migration circuits. White and grey shading indicate spatial separation of individuals between life history stages; graded shading indicates mixing of individuals. Diagram concepts are adapted from Smedbol and Stephensen (2001) and Secor (2010).

A contesting explanation to closed life cycles due to imprinting posits that social transmission of migration contributes to persistence in life cycles (McQuinn, 1997). This Adopted Migration (or Entrainment) theory postulates that life cycle circuits are learned by juveniles through their association with larger, more experienced individuals during periods of spatial overlap (Figure 2.4). Behavioral propensities to learn and school cause individuals to adopt migrations and life cycles through their interactions with older, experienced individuals who themselves adopted the migration pathways of previous generations. The theory holds that there is a labile period during which interactions with experienced individuals in schools or other aggregations lead to entrainment of individuals into a life cycle (Cushing, 1962; Guttridge et al., 2009). Such life cycles can persist over generations, causing individuals to return to natal habitats and eventually lead to reproductive isolation and associated genetic drift.

2.8 Self-Recruitment in Reef Fishes

For many reef fishes, natal homing is viewed from the perspective of a biphasic life cycle: (1) larval dispersal and (2) juvenile-adult residency. After the newly transformed juvenile has settled onto a reef, it remains there as an adult (or in the immediate vicinity) and does not undertake later migrations that would cause it to reproduce elsewhere. Reefs often occur in open advective environments, and lacking an adult homing migration, the presumption has been that most larvae will ultimately reproduce on non-natal reefs after dispersing some distance. This was tested in fringing reefs of Lizard Island, part of the Great Barrier Reef, where c. 10 million damselfish embryos were marked by short-term incubations in tetracycline, which permanently stained their otoliths (Jones et al., 1999). Using classic mark-recapture estimators, the small number of positive recaptures (n = 15) scaled up to rates of juvenile return to Lizard Island, termed self-recruitment, between 15% and 60%. Substantial proportions of self-recruits have now been demonstrated for other reef fishes and invertebrates with dispersive pelagic larvae (Cowen and Sponaugle, 2009).

2.9 Open Populations

The synergism of imprinting and straying represents a key emergent property of complex life cycles. Imprinting as a mechanism of natal homing would lead to maladaptive rigidity if not complemented by straying (Harden Jones, 1968). One can easily conceive how climate change, tectonic events, and other ecosystem changes could lead to extirpation of populations that relied upon a single rigid life cycle. In tandem, natal homing and straying provide a powerful synergism permitting marine fishes to capitalize on novel reproductive habitats within a single generation (Cury, 1994). Straying represents a form of ecological overhead: manifestly wasteful because in many instances mating by strays will often fail to produce progeny. Still, in some instances straying fish will successfully colonize new habitats, and in these instances, imprinting leads to fixation by progeny to a novel life cycle. Thus, imprinting causes conservatism that allows populations to capitalize on long-term stability in conditions that favor replacement. In this dual system of inertia and innovation, we might expect rates of natal homing among populations to relate to the stability of habitats required for offspring survival. Causes of straying remain largely speculative. Navigation error (overshoots, reverse migrations, imprinting errors) is a known source of straying for marine fishes and birds (Keefer et al., 2008; Newton, 2008). In well-studied Pacific salmon, straying is prevalent among populations and differs between sexes, indicating an underlying genetic propensity to stray (Quinn, 2005). Density is commonly invoked as a cause of increased straying particularly in metapopulations exhibiting source-sink dynamics (see Kritzer and Liu, Chapter 3, this volume). Due to the high fecundity of most marine fishes, even low relative rates of straying can produce substantial opportunities for strays to sample new environments.

The extreme cycles of population growth in marine planktivorous fishes such as California sardines or Peruvian anchoveta have long defied mechanistic explanations based upon climate, oceanography, food web, or fishery influences (Baumgartner et al., 1992). These population dynamics are commonly associated with a large change in spatial range, or in the parlance of avian ecology, irruptions (Schwartzlose et al., 1999; Newton, 2008; Petitgas et al., 2010). Bakun (2001) proposed that cycles of range expansion and contraction in many pelagic marine fishes correspond to abundance dynamics associated with predation and schooling. Periods of depressed abundance and contracted distribution occur due to efficient predation by co-occurring predators. However with decreased abundance, schools within a population become increasingly independent from each other. Then, a given school or school aggregation may adopt a more exploratory behavior and break out from its recent range, resulting in a release from predation and/or improved forage conditions. Increased growth, survival, and fecundity by this exploratory segment can stimulate a series of further range extensions by exploratory schools. As an example, Parrish and Edelstein-Keshet (1999) describe the rapid inundation of Fish Aggregation Devices (engineered flotsam set out to attract pelagic fishes) by large schools of tuna that rapidly overwhelm the initial attraction radius of the device. At some point, Bakun (2001) suggests that school aggregates merge to become a homogeneous hyper-school, resulting in the loss of previous migration behaviors (Figure 2.3). When food web or environmental conditions no longer favor growth and survival, the hyper-school, now recalcitrant to change, collapses at a much higher rate than would have occurred had diversity in population segments been maintained.

The rapid expansion and contraction of Japanese sardine in the 1970s and 1980s represents a possible case study (Bakun, 2001). In the initial phase of low abundance, two population segments were recognized: a nearshore, fast-maturing (age 1) group and a migratory, slow-growing (age 2–3) oceanic group. Some unknown environmental event(s) stimulated a linked change in abundance and range expansion, resulting in an autocatalytic response in which the migratory group became overwhelmingly dominant. By the mid-1980s the population's range showed remarkable expansion beyond its typical historical distribution, extending into the Sea of Japan, the shelf waters of Kamchatka, and the Kuroshio Current, ∼3000 km beyond its normal range. Bakun (2001) speculates that this super-migratory population was recalcitrant to change when conditions no longer favored its new distribution. During the period of range expansion, sardines apparently abandoned former inshore areas, which may have contributed to its rapid collapse during the late 1980s and early 1990s (Watanabe, 2002).

Although the above conceptual model (termed the school-trap by Bakun, 2001) is based on circumstantial evidence, such models can lead to improved hypotheses, broadened perspectives, and new avenues of research. In this case, intuitively we should expect schooling to influence broad-scale migration patterns and range dynamics of mobile planktivorous fishes, yet very few conceptual models link the two (Fréon et al., 2005). Bakun introduces schooling as a propensity for self-organization within fish populations. As defined by Ulanowicz (1997), this propensity does not represent an efficient (bottom up) mechanism of biological response, but rather a system with its own internal dynamics, which in this case is periodically reset by climate, fishing, and other external forces.

2.10 Between Closed and Open Populations: Connectivity

Connectivity, broadly defined as the exchange of individuals within networks of local populations or habitats, has emerged as a field of inquiry across all classes of migrating animals. The related concept of self-recruitment through limited larval dispersal, if applied more broadly across marine fishes, leads to well-trod ground: the migration triangle, hydrographic containment, and the member-vagrant hypothesis. The recent challenge is to recognize the open nature of populations: (1) how populations and groups exhibit both connected (i.e., correlated) and independent dynamics; (2) how populations are internally structured in ways that contribute to connectivity; and (3) what motivates threshold changes in population integrity that lead to irruptions, collapses, and shifts in ranges (Secor, in preparation). These issues of closed and open populations require perspective beyond the traditional emphasis on larval connectivity but an emphasis on all life cycles (Pineda et al., 2007). Several examples follow.

Interacting spawning groups of Atlantic herring exhibit varying amounts of covariation in their response to environmental conditions. Both internal dynamics within these groups and their aggregate responses to environmental forcing will be influenced by the degree and type of population connectivity. For a simulated two-component herring metapopulation, Secor et al. (2009) evaluated the emergent properties—yield, stability, and persistence—under connectivity scenarios of straying and behavioral entrainment. Under cycles of environmental forcing, density-dependent straying tended to stabilize populations while density-dependent entrainment often led to extirpation of one of the spawning groups. Higher levels of connectivity, regardless of type, tended to increase the degree of synchrony between the two groups and decrease metapopulation yield and stability. Most types of connectivity caused dominant year-classes to be distributed across the groups, destabilizing individual populations that would have otherwise benefited through the accumulation of spawning stock biomass (aka the storage effect; Warner and Chesson, 1985; Secor, 2007). Thus, the simulated herring metapopulations required some degree of connectivity for long-term persistence, but this may involve a tradeoff in terms of lost yield and stability. Results of the Atlantic herring simulation indicated that small levels of connectivity (∼5%) caused negligible influence, but at higher connectivity levels the internal dynamics of subgroups were disrupted, causing the entire metapopulation to destabilize.

Some stocks are structured as contingents, where partial migration, differential migration, entrainment, and other behavioral mechanisms cause groups within populations to undertake distinct lifetime migration behaviors (Secor, 1999; Chapman et al., 2011). In the Patuxent River (Chesapeake Bay) white perch population, spawners of resident freshwater and dispersive brackish water contingents fully comingle, but their progeny adopt contingent behaviors based upon larval growth conditions; thereafter they exhibit unique spatial behaviors and vital rates (Kraus and Secor, 2004; Kerr and Secor, 2009). Thus this estuarine-dependent population of white perch is structured according to both open (common reproductive pool) and closed (contingent-specific demographic trajectories) dynamics. In simulation studies, Kerr et al. (2010a) showed that long-term population yield, stability, and resilience depended on the degree of correlated response between the two contingents. Yield and stability increased when contingents negatively covaried in their response to external forcing. Empirical evidence supported separate dynamics where the resident contingent and dispersive contingents were respectively favored during drought and wet conditions (Kraus and Secor, 2004, 2005).

Atlantic bluefin tuna historically ranged between Brazil and Norway. Indeed important historical fisheries occurred off Brazil and Norway in the 1960s but abruptly disappeared in the 1970s. The fisheries off Norway and other parts of Northern Europe were historically important in the early twentieth century (MacKenzie and Myers, 2007). Fonteneau and Soubrier (1996) speculated that these historical fisheries occurred on unique segments of the population. These frontier contingents may have been lost initially because they were more vulnerable to the effects of exploitation. Once lost, conservatism of migration pathways (Corten, 2001) within other contingents could have precluded reinvasion of these historical feeding areas. This contention is supported by the absence of temperature and other environmental changes since the fisheries decline that would have been consistent with a shift in range (MacKenzie and Myers, 2007). Thus, range contraction by the species may have been due to loss of specific contingents that were maintained through entrained migratory pathways (Petitgas et al., 2010).

2.11 What Do We Need to Know to Track Fish Stocks?

The traditional and still widely held view is that stocks are reproductively isolated populations with their own internal dynamics, which can be identified by genetic markers of lineage (Booke, 1981; Ihssen et al., 1981). Molecular markers now support many scales of inference: phylogeny, evolutionary lineage, genetic drift, recent allopatric separation, natal and brood origin, and within generation selection (Waldman, 2005; Nielsen et al., 2012). On the other hand, assumptions in using these markers to differentiate marine populations remain largely unverified. These include whether alleles are neutral in their action (Gauldie, 1991), baseline rates in molecular clocks (Moritz, 1994; Drummond et al., 2006); incidence of hybridization (Hailer et al., 2012), Hardy–Weinberg population equilibrium (Ruzzante et al., 1996), the role of epigenetic inheritance (Bird, 2007; Navorro-Martin et al., 2011), and sampling error (Balloux and Lugon-Moulin, 2002; Dannewitz et al., 2005; Waples et al., 2008). The lack of a global molecular marker to determine reproductive isolation (Waples et al., 2008) has led to an interdisciplinary approach where advances in molecular approaches are tied with other nonmolecular approaches in identifying stocks and population units. These are given state-of-the-art reviews in this volume and include: (1) phenotypic (Chapters 6–9), demographic (Chapters 4, 5), and natural marker (Chapters 10–14) properties of stocks; (2) analysis of tagging (Chapter 16), telemetry (Chapters 17, 19), oceanographic (Chapter 15), and catch statistics; and (3) movement models (Chapter 18). When integrated, multiple approaches provide a weight of evidence framework in the practical identification and delineation of stocks (Cadrin et al., 2005; Waldman, 2005; Waples et al., 2008, Chapters 20–21).

Recent agendas to rebuild depleted and collapsed stocks, conserve endangered species, and protect critical habitats have expanded the traditional definition of stock to consideration of structural and behavioral entities within populations (e.g., subpopulations or contingents). Such structures can confer important yield, stability, and resilience functions (Secor, 1999: Ruzzante et al., 2006; Kerr et al., 2010a,b). In a series of case studies on collapsed and depleted marine fish populations, Petitgas et al. (2010) argued that recovery strategies that focused exclusively in rebuilding biomass in overfished populations were likely to fail. Rather, the spatial and behavioral architecture of those populations must be recovered as well. Conserving biomass among multiple components serves as bet-hedging against catastrophic losses of entire stocks (Smedbol and Stephenson, 2001; MacCall, 2012) and resilience of populations to climate change (Kerr et al., 2010a). Advances in ocean observing systems, telemetry, and molecular and other natural tags indicate that these stock identification tools can identify key structural/behavioral entities in support of conservation measures, but this will require expanded and costly research and monitoring programs.

The combination of traditional and more resolved analyses of stock structure has resulted in a burgeoning scientific literature, but in most instances stock structure information remains insufficient to support conservation aims. In 2011, 84 papers were published related to stock structure or stock identification, and over 2400 past papers on these themes were cited; papers in this field are well cited (h-index = 60; Web of Knowledge ©Citation Report, May 2012). However, studies are typically expensive, requiring intensive scientific efforts outside the scope of routine stock assessments, demands that often exceed the resources and capacities of governments and other stakeholder groups (Cope and Punt, 2009; Quetglas et al., 2012). Pertinent questions arise: What is the minimum amount of information we need to know to identify stocks and unit stock areas? How often do we need to test for changes in stock structure? Can stock structure be routinely tested in stock assessments?

More accessible approaches to evaluate stock structure in data-limited instances include improved delineations of harvest stocks and simulation models. Cope and Punt (2009) proposed that stock structure might be tracked efficiently through the analysis of covariance in spatially explicit abundance indices. Here, correlated dynamics are expected in instances of strong connectivity and/or strong demographic covariance across regions (Kraus and Secor, 2005; Rothschild, 2007; Manderson, 2008). Another approach is to ask through simulation models, when does stock structure matter? Management strategy evaluation is a simulation framework where alternative operating models (premises) are tested against their influence on assessment and management outcomes. In age-structured simulation models, Kerr et al. (2010b) showed that mis-specifying stock structure in Gulf of Maine cod could lead to underestimates of productivity. Here, independent dynamics of coastal subpopulations contributed to higher yield in the overall shelf metapopulation. In a similar exercise, Kell et al. (2009) observed that the consequences of lumping rather than splitting population components of British Isles herring caused a virtual population assessment model to yield optimistic predictions on the level of overall fishing rates and probability of recovery following depletion. An advantage of simulation modeling is the flexibility in accommodating multiple types of information and levels of structural organization (Kerr and Goethel, Chapter