Assessing Ecological and Economic Effects of Derelict Fishing Gear: a Guiding Framework Christopher F. G. Jeffrey1, Kirk J. Havens2, H. Ward Slacum, Jr., Donna Marie Bilkovic2, Danielle Zaveta3, Andrew M. Scheld2, Sean Willard1,John D. Evans4 Revision 1 - Oct. 7, 2016 Prepared for Marine Debris Program Office of Response and Restoration National Oceanic and Atmospheric Administration Contract DG133E-10-CQ-0034, Task Order 007 1CSS-Dynamac, Inc. 2Virginia Institute of Marine Science Derelict Fishing Gear Assessment Framework 3Versar, Inc. 4Global Science & Technology, Inc. 1 Derelict crab pot removal in the Chesapeake Bay (Photo credit: Matt Stewart) Assessing Ecological and Economic Effects of Derelict Fishing Gear: a Guiding Framework Revision 1 – Oct. 7, 2016 Table of Contents Executive Summary ..........................................................................................................................................ii 1. Introduction / Background .....................................................................................................................1 1.1 Background – NOAA’s Marine Debris Program..........................................................................1 1.2 The Problem of Derelict Fishing Gear ..........................................................................................1 1.3 Purpose and Scope ...........................................................................................................................3 1.4 Intended Audience ...........................................................................................................................3 2. Derelict Fishing Gear Assessment Framework ...................................................................................5 3. Characterizing Derelict Gear Density and Spatial Distribution ......................................................8 3.1 Overview .............................................................................................................................................8 3.2 Use of Various Types of Datasets .................................................................................................8 4. Assessing Ecological Effects of Derelict Gear .....................................................................................8 4.1 Overview .............................................................................................................................................9 4.2 Mapping / Quantifying Bycatch Species ......................................................................................9 4.3 Quantifying Mortality Rates ........................................................................................................ 10 4.4 Mapping Sensitive Habitats ......................................................................................................... 11 5. Assessing Economic Costs of Derelict Gear ..................................................................................... 12 5.1 Overview .......................................................................................................................................... 12 5.2 Characterizing Fishing Effort and Harvest .............................................................................. 12 5.3 Decreased Harvests....................................................................................................................... 12 5.4 Habitat Damage ............................................................................................................................. 13 5.5 Navigational Hazard ..................................................................................................................... 14 5.6 Replacement Gear ......................................................................................................................... 14 6. Detecting and Removing Derelict Fishing Gear ............................................................................... 15 6.1 Overview .......................................................................................................................................... 15 6.2 Derelict Gear Detection ................................................................................................................ 15 6.3 Derelict Gear Removal.................................................................................................................. 16 7. Evaluating Management Implications ................................................................................................ 18 7.1 Spatial and Temporal Management of the Fishery ................................................................ 18 7.2 Data Reviews, Compilation, and Synthesis .............................................................................. 19 7.3 Data Collection Plans to Fill Critical Data Gaps ..................................................................... 19 Appendix A. Global Derelict Fishing Gear Estimates ............................................................................ 20 References ....................................................................................................................................................... 22 Acknowledgments ......................................................................................................................................... 27 i Executive Summary Developing standardized protocols to assess the ecological and socio-economic effects of marine debris – especially, derelict fishing gear – is critical for the protection of natural resources and for evaluating policies and programs designed to reduce and remove debris. This document outlines a Derelict Fishing Gear Assessment Framework to guide the development and implementation of derelict gear assessment, management and mitigation. The framework draws from techniques and protocols developed to assess derelict crab traps effects in the Chesapeake Bay and on past derelict gear assessments either conducted by or known to the framework authors. However, this framework is generalized and intended to be used by any stakeholder with a need to assess the status of derelict fishing gear and its economic and ecological effects on living resources, habitats, ecosystems, and local economies. It provides a generalized pathway and processes for assessing the effects of derelict fishing gear, and is flexible and scalable so that users of the framework can make informed decisions when data are limited, and can tailor it to satisfy their specific assessment goals and objectives if a full scale assessment is not required. The framework recommends best practices for each of five key elements (Figure 1):  Characterize the abundance and distributions of derelict gear and associated fisheries;  Conduct analyses to quantify economic impacts on fishery target and bycatch species;  Conduct analyses to determine ecological impacts on living resources and critical or sensitive habitats;  Evaluate management implications through scenario driven economic and ecological assessments;  Recommend appropriate management actions and strategies to mitigate negative economic and ecological impacts of derelict gear. Descriptions of appropriate derelict gear metrics are provided with an emphasis on those variables and metrics most common among fisheries, and most likely to predict ecological or economic effects. For each of the key elements, the framework provides guidance on consistent methods for stakeholder engagement, data collection or acquisition, and data analysis. Section 1 of this document provides a global context and motivation for assessing and managing derelict fishing gear. Section 2 outlines the framework and essential ingredients including developing conceptual models, planning an analytic approach, evaluating data, and developing quantitative models. Sections 3, 4, and 5 describe technical details and procedures for characterizing the density and spatial distribution of active and derelict fishing gear; for mapping and quantifying vital rates to assess ecological effects of derelict gear; and for determining data needs and approaches to assess economic effects of derelict gear on fisheries. These sections emphasize detailed guidelines on required data, information, methods, approaches, protocols, as well as best practices and data gaps for each element in the framework. Each section includes descriptions of common data, data limitations, modeling, and other analytical techniques. Derelict Fishing Gear Assessment Framework ii Figure 1. Conceptual assessment framework to characterize, assess, and mitigate economic and ecological impacts of derelict fishing gear on fisheries, living resources, habitats, and ecosystems. Section 6 provides an overview of techniques used to detect and remove derelict fishing gear, so as to facilitate the collection of data needed to assess their effects on local ecosystems and economies. Many techniques to detect and remove derelict fishing gear are available and have been applied in various fisheries and habitats. A list of specific techniques and references is provided here, along with specific protocols to help guide stakeholders interested in collecting needed data to characterize derelict gear or to mitigate their effects through detection and removal programs. The final framework element (Section 7) describes how a structured derelict gear assessment can be used to inform resource managers about mitigation options. This section provides guidance of how results from derelict fishing gear assessments can be used to relate the ecological and economic effects of derelict gear to management implications and understanding the factors affecting implementation of mitigating actions. The ultimate goal of any derelict fishing gear assessment should be to determine how best to manage and mitigate derelict gear effects. This Derelict Fishing Gear Assessment Framework is designed to be a flexible, scalable, practical guide to help stakeholders meet this goal. Derelict Fishing Gear Assessment Framework iii 1. Introduction / Background 1.1 Background – NOAA’s Marine Debris Program Marine debris (including derelict fishing gear) is a major stressor to marine Summary of Relevant Goals from the NOAA ecosystems; it threatens habitats, living Marine Debris Program Strategic Plan resources, and the provisioning of  Marine Debris Research & Assessment - Identify, ecosystem goods and services in U.S assess and reduce the impacts of marine debris waters and the global oceans.1 The through detection, monitoring, source Marine Debris Act, signed into law in identification and innovative solutions. 2006 and amended in 2012, defines  Marine Debris Removal - Support the removal of marine debris as “any persistent solid marine debris to reduce its impact on natural material that is manufactured or resources, navigational safety and the economy.  Marine Debris Prevention - Prevent the processed and directly or indirectly, introduction of marine debris through raising intentionally or unintentionally, awareness and changing behavior. disposed of or abandoned into the NOAA Marine Debris Program Strategic Plan: 2016 - 2020 marine environment or the Great (October 2015) Lakes.” The Marine Debris Act establishes the NOAA Marine Debris Program “to identify, determine sources of, assess, prevent, reduce, and remove marine debris and address the adverse impacts of marine debris on the economy of the United States, the marine environment, and navigation safety.”2 1.2 The Problem of Derelict Fishing Gear The accumulation and effects of derelict fishing gear (traps, pots, nets) is a global problem.3-6 Traps are fished in large numbers and high rates of trap loss plague many of the world’s crustacean fisheries (Appendix A, Table A-1; Figure 2); as a result, fishing traps are thought to be one of the most common types of derelict gear worldwide.7 This framework focuses on traps and stationary nets (i.e. gill nets). Derelict fishing traps, whether lost by accident or intentionally discarded, have a tendency to continually capture animals (termed “ghost fishing”) for variable amounts of time depending upon the trap construction and material type.4,9-13 Modern traps are often constructed from rigid and durable materials such as vinyl-coated wire or synthetic mesh6,14,15 which can extend the time a derelict trap remains functional over the more artisanal traps of the past which were made mostly of wood or fiber.16-20 Experiments in the Chesapeake Bay (U.S.A.) indicate that blue crab (Callinectes sapidus) traps generally retain their structural integrity for two years or more.1,12 Derelict traps can act as a structural attractant for crustaceans or an aggregating device for fish.21-25 Blue crabs and other crustaceans are known to be attracted to traps as bottom structure whether or not any bait is Derelict fishing gear can continue to capture present; with retention rates varying animals and is sometimes referred to as “ghost according to trap design and intra- and interfishing.” Some derelict traps have been shown species interactions.4,21-23,26 Animals that are to continue functioning for up to 15 years. captured and die in derelict traps can attract Derelict Fishing Gear Assessment Framework 1 other animals, which then become entrapped and continue a ‘self or auto-baiting’ cycle.12,13, 27 Catch efficiencies of derelict traps can decline over time;10,28 however, many trap types can continue to capture, injure, and kill animals for over two years 4,12-14,29,30 and in some cases can remain functional for at least 15 years.31 More durable traps made of predominantly synthetic material may cause ecological and economic damage for many years.32 In addition to the direct loss of target species in Derelict fishing gear can affect harvest by: derelict traps due to mortality, derelict traps  Direct mortality of target species impose an economic cost in terms of reduced  Competition with active gear for target trap efficiency by competing with active traps species. for the target species resulting in a hidden nonharvestable allotment of target species associated with derelict traps. This loss of harvest can compromise the economic vitality of fishery dependent businesses and communities.8 In the Chesapeake Bay, extensive gear removal programs between 2008 and 2014 increased blue crab harvest by 38 million pounds (23.8%) over the six year period.33 For the U.S. Atlantic and Gulf state blue crab fisheries, extensive gear removals might increase landings by over 40%, generating US $62 million in annual revenue benefits. However, large-scale location and removal of submerged, unbuoyed traps requires expensive equipment and may not be practical over large areas or for extended periods. Fortunately, intensive (focused) removals of derelict traps from a few heavily fished areas can also increase local harvests. In the Chesapeake Bay, removing as little as 10% of the total number of recovered derelict traps from the 10 most heavily fished areas (five in Virginia and five in Maryland) increased the Bay-wide blue crab harvest by 14%.33 Figure 2. Global distribution of major crustacean pot and trap fisheries. 8 Total global landings from all crustacean trap fisheries grossing US $20 million or more annually (Figure 2) average 615,560 MT and are worth US $2.5 billion (Appendix A, Table A-2). Together, these high-value fisheries deploy tens of millions of pots and traps, millions of which become derelict each year. Extending findings Removal of just 10% of derelict traps in the from Chesapeake Bay blue crab to global crustaworld’s largest crustacean fisheries could cean fisheries suggests that removing less than increase harvest by $831,000,000 annually. 10% of the derelict pots and traps in these Derelict Fishing Gear Assessment Framework 2 fisheries could increase landings by 293,929 MT, at a value of US $831 million annually. 8,33 Net benefits of removal programs will ultimately depend upon trap location and removal costs however, which may vary widely. 1.3 Purpose and Scope Derelict fishing gear is a major concern for coastal resource managers, given its widespread occurrence and its far-reaching negative ecosystem impacts. To guide characterizations and assessments of the impacts of derelict fishing gear on marine ecosystems, and the fisheries they support, we propose a Derelict Gear Assessment Framework as a guide for local assessment and management of derelict fishing gear problems. It provides a generalized but scaled approach for assessing derelict fishing gear impacts; describes requisite data and information needs for a comprehensive assessment of ecological and economic impacts; and where possible, recommends best practices and management actions to remove and prevent accumulation of derelict fishing gear and ultimately to mitigate economic and ecological impacts. This Framework is modeled after a recent comprehensive characterization and assessment of derelict crab traps in the Chesapeake Bay region.33 It is applicable to the assessment of derelict fishing gear impacts in general and provides a standardized approach for managing derelict gear and mitigating their impacts on ecosystems nationally. 1.4 Intended Audience This Framework is primarily intended for NOAA Marine Debris Program managers, policy makers, natural resource managers, and others who perform work or receive sponsorship under NOAA’s Marine Debris Program or follow its guidelines and regulations concerning marine debris. The style, language, terminology and concepts described here are intended to be useful to a diverse audience including Federal, State, and local agencies, scientists, members of the general public, and other stakeholders who may need to characterize marine debris or assess its economic and ecological impacts on living resources, habitats, ecosystems, and local economies. We hope that this report will be a first step in developing a broad scientific consensus about assessments of derelict gear. Derelict Fishing Gear Assessment Framework 3 Chesapeake Bay Case Study The Chesapeake Bay Blue Crab fishery. The guiding framework detailed here was initially developed for and 33 applied to the Chesapeake Bay , the largest estuary in the United States with a fishing industry worth more than $3 billion a year and accounting for 50% or more of the nation’s blue crab (Callinectes sapidus) harvest. The primary gear used to harvest blue crabs in the Chesapeake Bay is a rigid cube-shaped wire trap, or “crab pot,” galvanized or vinylcoated, and deployed and recovered by a line and buoy system. Commercial fishery effort is estimated to be ~600,000 34 pots annually. Blue Crab management: one population, three sets of rules. Chesapeake Bay blue crab is a single population, but it is managed differently in the Maryland, Virginia, and Potomac River portions of the Bay. Management actions are partially linked to the complex life cycle of the blue crab, which involves multiple horizontal and vertical migrations, often across state boundaries. Blue crabs rely on high salinity areas in the lower Bay for egg brooding and hatching, larval development, and overwintering;35-37 they inhabit low salinity waters of tributaries and creeks in the spring and summer;35 and when temperatures drop below 9°C they congregate in Figure 3. Spatial distribution of derelict crab pots deeper water and bury in muddy sediments.38-40 Spatial patterns of fishing effort follow the crabs’ migration; most crab in the Chesapeake Bay estimated through 33 geographically weighted regression. pots are deployed in waters less than 10m deep, in the mainstem of the Bay and (in Virginia) in its tributaries.41,42 Data collection. In order to evaluate the ecological and economic impacts of derelict crab pots, information was first compiled to determine 1) the distribution and abundance of derelict pots, 2) bycatch capture and mortality, 3) potential damage to sensitive habitats, and 4) the effect on harvest due to competition with derelict traps. After learning of potential causes of pot loss from fishers and managers, the distribution of derelict pots was compared statistically with these potential causes including boat traffic, storm events, and intentional abandonment. Data analysis. The study integrated many datasets using a Biogeographic Assessment Framework 43 to  identify variables that could predict the distribution and abundance of derelict crab pots,  model, estimate, and map the densities and spatial distributions of derelict crab traps, and their impacts on populations of crab and other fish and on sensitive habitats;  estimate potential impacts of derelict crab trap removals on commercial blue crab harvests;  conduct sensitivity analyses to identify ways to mitigate the impacts of derelict crab pots. Findings and insights.  Trap loss was due primarily to boat traffic (severing buoy lines, displacing gear) and abandonment (due to aging gear, sickness or death of fisher, other).  Using fisheries independent effort data led to better estimations of derelict fishing gear abundance in areas with large management units such as the Maryland mainstem portions of the Bay.  Each year, derelict pots kill over 3.3 million blue crabs across the Chesapeake Bay (4.5% of the harvest).  12-20% of crab pots are lost each year; this implies over 145,000 derelict pots in the Bay at any given time.  Over 40 fish species have been reported captured in derelict pots, including economically important ones. Each year derelict pots kill over 3.5 million white perch and nearly 3.6 million Atlantic croaker.  Extensive gear removal programs spanning 2008-2014 increased blue crab harvest by 38 million lbs over the 6 year period. In addition, targeted gear removals from a few high-density fishing areas disproportionately increased harvests Bay-wide. Management actions.  Minimize boat traffic in crabbing areas, and educate boat operators about avoiding pots to reduce pot loss.  Conduct targeted derelict pot recovery programs to reduce bycatch mortality and to increase catch efficiency from lost pots. Also incentivize the removal of abandoned pots by fishers themselves.  Equip crab pots with biodegradable escape panels to minimize bycatch mortality from lost pots. Derelict Fishing Gear Assessment Framework 4 2. Derelict Fishing Gear Assessment Framework Like traps, other fishing gear affects both the living resources being targeted as well as the surrounding habitats in which they live. In turn, the heterogeneity of the seafloor and habitats, patchy distributions of targeted living resources, and environmental factors such as seasons, storms, and wave action, all affect the location and accumulation of derelict gear. Additionally human demography, behavior, as well as availability of access to the marine environment directly affect the location and intensity of fishing effort, which ultimately determines the occurrence of derelict gear. Developing strategies to manage derelict gear requires a comprehensive approach for characterizing the spatial and temporal distributions of derelict gear, assessing their economic and ecological effects, and implementing management strategies to mitigate their negative impacts on living resources, habitats, and the marine ecosystems. The Framework presented here represents such a comprehensive approach. It provides a fivephased linear process model for characterizing, assessing and mitigating ecological and economic impacts of derelict fishing gear on managed living marine resources, habitats, and ecosystems (see Figure 1, Executive Summary). It identifies generalized pathways and processes needed to: 1. Characterize spatially explicit abundance and distributions of derelict gear and their associated fisheries 2. Conduct analyses to quantify economic impacts on fishery target and bycatch species 3. Map and characterize benthic habitats then conduct analyses to determine ecological impacts on living resources and critical or sensitive habitats 4. Evaluate management implications through scenario driven economic and ecological assessments 5. Recommend for implementation, appropriate management actions and strategies to mitigate negative economic and ecological impacts of derelict gear Subsequent chapters provide more detailed guidelines on required data, information, methods, approaches, protocols, as well as best practices and data gaps for each phase of the framework. In addition, the Framework provides a tiered approach which allows for informed decisions based upon various levels of available data and to adaptively manage impacts of derelict fishing gear based upon the particularities of the specific fishery of interest. Developing a comprehensive framework to manage derelict gear and mitigate their impacts involves developing conceptual models, designing an appropriate analytic approach, evaluating data to determine minimum requirements, and developing quantitative models. Conceptual models. An important first step is to develop conceptual models based on sound ecological theory but which are also grounded in socio-economic reality (Figure 4). Conceptual models are hierarchical theoretical constructs visualized in flow diagrams to show qualitative relationships among components that represent dependent and independent variables within complex systems. These models are widely used in systems ecology to hypothesize about direct and indirect relationships among various biotic and abiotic elements within a defined and bounded system.44,45 A conceptual model will provide a systematic approach for (a) identifying components of the ecosystem that could be directly or indirectly affected, (b) evaluating their relative contribution to hypothesized ecological and economic effects, and (c) hypothesizing further about possible ways to mitigate negative impacts. Derelict Fishing Gear Assessment Framework 5 Figure 4. An operational (tiered and practical) assessment framework for derelict fishing gear. Critical / minimum data on derelict fishing gear (DFG) (left panel) enable analyses, metrics, and indicators (middle) required to mitigate DFG economic and environmental impacts for long-term sustainability (right). It is important then that conceptual models be based on exhaustive reviews of the scientific literature on fisheries and derelict gear. In addition, input from local experts, non-scientists, and fisher-folk who are most familiar with local fishery issues will be substantially beneficial. Design the Analytic Approach. Designing and planning a suitable analytic approach is critically important for developing a framework. Planning is useful, not only for developing higher level project goals and objectives, but also for developing specific guidelines for conducting quantitative data analyses, ensuring the use of statistically defensible analyses, and increasing the likelihood of achieving unambiguous results. Given the paucity of quantitative information regarding ecological impacts of derelict gear, it is essential to develop simple but realistic hypothesis-driven analyses that will support the higher-level goals and objectives of a derelict gear impact assessment project. At the same time, question driven analyses regarding derelict gear impacts should be developed within the broader context of the socio-economic, ecological, and management issues related to the fishery to ensure the most critical and influential factors are being considered. Planning involves iterative refinement of the hypothetical relationships among conceptual model components as well as adding or eliminating input variables based on initial exploratory analyses of data, additional literature review, and consultation with management agencies, NGOs, and other stakeholders. During the planning stage, it is also necessary to develop a Derelict Fishing Gear Assessment Framework 6 spatial framework for analysis that defines both the spatial extent of the study area and the resolution of the data (i.e. level of detail) needed to characterize derelict gear and quantify ecological and economic impacts. An output from sound planning will be the identification of a reduced list of input variables that correlate significantly with dependent variables and which could be used as predictors of dependent variables that are sparse or expensive to obtain or collect (Figure 4). Ideally, input variables should be ubiquitous with broad spatial and temporal coverage and correlate significantly with dependent variables of interest that are sparse. Evaluate Data and Determine Minimum Requirements. Obtaining primary data on the occurrence and distribution of derelict fishing gear can be prohibitively expensive. Hence, relevant data regarding the types, distribution, occurrence, and ecological or socioeconomic impacts of derelict gear is unavailable in most fisheries. Except for a handful of targeted derelict fishing gear removal programs (such as exist in the Chesapeake Bay), much of the existing knowledge regarding derelict gear and their impacts derive from secondary data sources and disparate programs designed to collect other fisheries-related or ecological data. Consequently, data to be used for characterizing derelict gear occurrence and distributions and inferring ecological impacts must be carefully evaluated for their suitability to these intended purposes. Evaluating data to assess derelict gear impacts includes determining the best and most appropriate data sources; assessing each dataset’s geographical, temporal, and compositional extent and associated gaps; and understanding the uncertainty, errors, and associated caveats of the data. Crucial considerations for evaluating data include the following:     Are the data of appropriate spatial and temporal coverage? Do datasets contain enough observations for robust statistical analyses? Can disparate datasets be merged for broader spatial and temporal coverage without violating the statistical assumptions underlying the initial data collections? Will extrapolations of integrated data across geographic space and time provide ecologically meaningful relationships rather than spurious associations? Develop Quantitative Models. Model development involves the use of empirical data to test and validate theoretical constructs and hypothesized relationships through explicit mathematical functions. Quantitative models consist of three main parts: a response variable whose value changes as observational data inputs into the model change, a mathematical function with main effect variables and estimated parameters that predict the value of the response variable, and random errors that vary systematically according to known statistical distributions. Modeling requires a fundamental understanding of how the main effects and errors are propagated through the modeling system. A wide variety of modeling approaches are available for use; all share three common steps: (1) Selecting an initial model form that would be best suited to the data assembled; (2) Iteratively fitting the model to the assembled data to estimate unknown parameters; and (3) Independently validating the model results against known data to see if the underlying model assumptions are plausible. Sections 3, 4, and 5 below illustrate the iterative use of various modeling techniques to characterize derelict gear density and spatial distribution; to assess ecological impacts to bycatch and sensitive habitats; and to determine economic impacts to the fishery. Derelict Fishing Gear Assessment Framework 7 3. Characterizing Derelict Gear Density and Spatial Distribution 3.1 Overview Characterizing the spatial distribution of derelict fishing gear is crucial for inferring their ecological and economic impacts in marine ecosystems. The accumulation of derelict fishing gear in many aquatic systems varies spatially and temporally. To obtain unbiased estimates of the quantity and spatial coverage of derelict gear within an area of concern, it may be costeffective to model gear abundance and distribution, especially for large areas where comprehensive detection surveys and removal programs may be too expensive. When developing a model to estimate abundance and occurrence, it is important to select a suitable spatial resolution that matches the scale at which ecological inferences will be made and at which management decisions to mitigate derelict gear impacts will be effective. Often, selection of an optimal spatial resolution for analysis is constrained by the resolution of data available and depends on the analytic specific analysis being conducted or question being investigated. Derelict gear estimation may require further data collection to refine the spatial domain and extent of analysis. A variety of approaches can used to estimate gear abundance and occurrence, however the method and data used for these estimates will have different biases and limitations. 3.2 Use of Various Types of Datasets Many types of fisheries-related information may be used for initial characterization of derelict gear. For example, fisheries effort data on the amount and location of active gear sets, combined with reports from fishers about their average rate of gear loss, could be used to derive estimates of derelict gear accumulation. Estimates from such data will be constrained by the size or spatial scale of the management units applicable to the fisheries data. Additionally, fisheries data and reports of gear loss from active fishers may be subject to various biases such as the number of individuals interviewed, regional differences, methods or techniques used by different fishers, or the fishery regulations under which they operate. For example, incentive programs that compensate fishers for lost gear may lead to overestimating rates of gear loss. Conversely, fishers may underreport effort data if they are subject to regulations that penalize violations of gear limits. Depending on the goal, this type of derelict gear estimation may nonetheless be adequate for a coarse understanding of amount and location of gear. Data on derelict gear occurrence can also be obtained from gear retrieval programs. These often have a limited spatial coverage because their implementation costs are relatively high; however georeferenced data from such programs can be used to estimate (predict) gear abundance over larger spatial coverages through geospatial modeling (below).33 However, it is important to understand potential biases inherent in this type of data. For example, interpreting and standardizing data from retrieval events requires estimating the retrieval effort: how did search efforts (number of hours, number of individuals involved) vary across the recovery area? Were there different techniques that would lead to different recovery efficiencies? etc. Furthermore, dataset normalization can help stem issues with false zeroes or seemingly low or high derelict gear abundances. So while overall concentrations and effects of derelict gear information can be Derelict Fishing Gear Assessment Framework 8 inferred from gear retrieval efforts, they are influenced by a wide range of factors. Before making broad spatial inferences, it’s important to have a clear understanding of these factors throughout the system, and to collect data from a range of areas. Depending on the amount and quality of data available, a wide range of spatial techniques may be used for estimating derelict gear quantity – from simple multiplication, kriging, and interpolations, to advanced spatial modeling. Modeling requires (i) an understanding of the factors that may lead to gear loss and accumulation, (ii) spatial data for those factors, and (iii) a known relationship between them. This relationship can be developed using training data or a set of locations with known derelict gear abundance along with corresponding factor data at those locations. A key advantage of this technique is being able to predict gear abundance at different locations when only condition or factor data is available. However, the model’s prediction power is only as strong as the modeling relationship between known derelict gear and contributing factors. As with many modeling techniques, it is important to limit predictions within spatial coverage of the training data and training factor values. 4. Assessing Ecological Effects of Derelict Gear 4.1 Overview Ecological effects of derelict gear can occur at a variety of spatial and temporal scales and through several pathways ranging from simple direct interactions with habitat and living organisms to more indirect but complex relationships among dynamic populations, sensitive habitats, and the fishery. For most managed areas and fisheries, sufficient data do not exist at the required spatial coverage or temporal and spatial scales to fully define the ecological relationships between derelict gear and the various components of the marine ecosystem. As such, the framework provides a systematic and scalable approach for integrating existing data and other information on derelict gear, target and bycatch species, and sensitive habitats to characterize and assess ecological impacts from derelict gear over broad spatial scales. The approach involves using GIS-based models to (1) predict densities and define the spatial distribution of derelict gear, (2) apply estimated derelict gear capture and mortality rates obtained from experimental studies to spatial distributions of target and bycatch species, and (3) define the spatial overlap between derelict gear distributions and sensitive habitats. 4.2 Mapping / Quantifying Bycatch Species Quantifying ecological impacts of derelict gear on bycatch species requires information on catch species composition, rates of gear capture and mortality, and spatial distribution of non target species. Experimental field studies can provide crucial and representative data on catch composition and mortality rates if they are conducted over extended periods and investigators track and record the condition of trapped individuals with enough frequency. However, field experiments require rigorous and time intensive field work that may be too difficult and expensive to implement over large areas and sample sizes. In addition to field experiments, valuable but conservative data on species composition, capture, and mortality from gear could be Derelict Fishing Gear Assessment Framework 9 obtained also from large retrieval programs that periodically remove gear, if the composition and condition of the catch is noted when gear is removed. One caution is that seasonal and spatial variability may prevent the use of incidental bycatch data in deriving annual catch and mortality estimates for derelict gear. In addition, observed estimates of mortality, even from experimental studies, may underestimate actual gear mortality unless delayed mortality from injury, stress, infection, and fatigue are also considered.6,10,44 Bycatch species in derelict gear may include both managed and unmanaged species. Economically important non-target species or threatened/endangered (T&E) species that are managed and/or well-studied may have available data on distributions, abundance, and harvest that can greatly help inform potential adverse effects from derelict gear including direct and indirect mortality or damage to essential habitat. There may also be enough of the requisite data available for non-target non-fisheries species that are well-studied. For example, Diamondback terrapin (Malaclemys terrapin) is a species that is at high risk to mortality from active or derelict crab traps.47,48 In many regions, there are limited data on population sizes for this nonfisheries species; however, the habitat requirements for terrapin are well understood and mapping important habitats could serve as a surrogate measure for potential risk to the population. This approach could also be used for other T&E species with historically low populations that make mortality estimates challenging. Those species that are not managed or well-studied may require additional directed studies to assess specific capture and mortality rates caused by derelict gear. 4.3 Quantifying Mortality Rates In order to establish a sustainable fishery, it is necessary to effectively manage the primary sources of fishing mortality, including losses due to derelict gear.49 Animals captured in derelict gear, including invertebrates, finfish, turtles, birds, and mammals can starve, cannibalize each other, drown, develop infections, and become diseased.4,10,48 Certain crustaceans are known to be aggressive and cannibalistic (e.g.. blue crab, Callinectes sapidus), and confinement in traps causes death or injury to conspecifics and bycatch.4,23,25 Mortality rates in derelict traps vary among fisheries. For example, in the Korean blue crab (Portunus trituberculatus) fishery, the elimination of ghost fishing mortality by derelict traps was modeled to have a 14% increase in catch50 while Dungeness crab mortality in derelict traps is estimated at 7%, 4.5%, and 2.2% of the harvest in various locations along the west coast of North America.27,28,30 In the Chesapeake Bay, mortality due to derelict crab pots is estimated at about 5% of the harvest.33 Cryptic or unaccounted mortality resulting from derelict gear has been increasingly recognized as a source of mortality that can be managed.6,33,49,51 Collecting essential baseline data for calculating instantaneous and annual bycatch mortalities by gear type is an important consideration for any derelict gear recovery program52 but adequate data is often lacking and, thus, is not routinely accounted for in fisheries management. Data on mortality rates require experimental studies and field observations. Ghost-fishing mortalities can be estimated either by monitoring the number and fate of animals that became captured, or by placing captive animals in the gear and monitoring their fate over time, with monitoring conducted in situ directly by divers and video surveys, or by retrieving a subset of the derelict gear at various time intervals.49 Various methods for deriving mortality per unit period of time for derelict traps and nets are discussed in some detail elsewhere.4,6,49,53,54 Derelict Fishing Gear Assessment Framework 10 4.4 Mapping Sensitive Habitats Benthic habitat and living resource maps often characterize and delineate the spatial extents of important ecological areas such as areas of high productivity and biodiversity (hotspots); locations of special features, communities and key species that are critical to ecosystem function and resiliency (e.g. submerged aquatic vegetation, coral reefs, or oyster beds); or rare, endangered or functionally vulnerable marine resources (e.g., sea turtles). When these maps are combined with georeferenced data on derelict gear occurrence, density, and distribution, the degree of spatial overlap between derelict gear and sensitive habitats can be measured and used as a proxy for estimating ecological effects at a broader ecosystem level. The most obvious ecological effect of stationary fishing gear is direct physical disturbance to the seafloor. In addition to direct observations of fisheries gear-habitat interactions from field and experimental studies, the cumulative ecological effects of derelict gear deployed over large areas can be estimated. If the locations and spatial extents of fishing gear and important habitats are known, then the degree of interactions between derelict gear and sensitive habitat could be inferred through geospatial modeling. In the U.S., large investments in mapping by government agencies (e.g., BOEM, NASA, NOAA, USGS) and the private sector (e.g., Google, ESRI), along with recent improvements in remote sensing technologies (e.g., satellite, lidar, hydroacoustic sonar) have created a plethora of high resolution (< 100m) benthic and living resource distribution maps for many managed areas. Where such maps are not available, qualitative data from fishers and other local experts on intensity of various human uses and location of targeted resources can be captured through participatory mapping and used to delineate perceived ecologically important areas. Important caveats are that maps should be comprehensive in their spatial coverage (i.e. fully represent the study area of interest), have suitable spatial resolutions, and have relatively high map accuracy (i.e. users must have a high degree of confidence in the map) if they are to be used effectively in identifying ecological effects over broad spatial scales. As an example, when modeled distributions of derelict pots in the Chesapeake Bay were intersected with distribution maps of two sensitive habitats (submerged aquatic vegetation and oyster beds), results suggested minimal spatial overlap between derelict gear and these habitats of concern.33 It is likely that blue crab operators generally may be avoiding crabbing in submerged aquatic vegetation and oyster habitats. However, anecdotal data from trap removal studies suggest that derelict traps do occur in some areas with submerged aquatic vegetation. One likely explanation is that National Environmental Policy Act requirements restricted derelict pot removal from sensitive habitats, and that model predictions of derelict trap spatial occurrence may have been underestimated. In addition, although extensive portions of the Chesapeake Bay have been mapped for sensitive habitats, spatial coverage of the maps is not 100%, and this could also have led to underestimating the degree of spatial overlap between sensitive habitats and derelict trap distributions. Derelict Fishing Gear Assessment Framework 11 5. Assessing Economic Costs of Derelict Gear 5.1 Overview Economic costs imposed by derelict gear can result through a variety of mechanisms, which may vary according to attributes of the fishery, ecosystem, and broader region under consideration. Major categories of costs include: decreased harvests, habitat damage, hazards to navigation, and replacement gear.33 To measure these costs, a number of different models can be applied to collected and available data. Assessing the economic impacts of derelict gear frequently requires comparing status quo economic outcomes and valuations against a hypothetical of no or minimal levels of derelict gear (i.e., what economic outcomes or values would be without derelict gear). This section discusses fisheries data and four categories of economic costs, together with general models and data needed for empirical estimation. 5.2 Characterizing Fishing Effort and Harvest Integral to understanding the distribution and the potential impact of derelict fishing gear is the knowledge of both fishery effort and fishery harvest. Fishing effort, or the measure of amount of fishing activity, is often estimated through the combination of effort put into the fishery such as the number of hours or days spent fishing and the amount of gear used. Harvest data is generally reported as the amount of the target species caught and retained from a given area for a given period of time. Both effort and harvest data can be collected from fishery-dependent sources (direct reporting from the fishery such as vessels or seafood dealers) or from fishery-independent sources. Fisheries scientists or managers often collect fishery-independent data while conducting resource-monitoring projects. Although fisheries-dependent data is often easier and cheaper to obtain, it is hard to determine its accuracy especially if there is skepticism for potential bias. As noted earlier (Section 3.2), special consideration should be taken to review all potential sources of bias when fisheries-dependent data is used and to plan for ways to mitigate the bias either through conversion or using different sources. Fishery-independent sources often provide less biased information and can be tailored to specific applications. Furthermore, effort data reported at the state level or for large management units may be too coarse to be used in impact assessments when derelict gear presence may vary at more fine spatial scales. Depending on the fishery and the spatial and temporal scale of the derelict gear assessment, it may be necessary to use fisheries-independent data to obtain the spatial and temporal resolution needed. 5.3 Decreased Harvests Derelict gear may decrease commercial and recreational harvests of target and bycatch species. Reductions in harvests arise as a result of reductions in stock as well as from the reduced efficiency of actively fished gear. Stock reductions may be significant in fisheries with high rates of capture and mortality of target and bycatch species by derelict gear. If derelict gear leads to large stock reductions, commercial and recreational harvests may also decrease as there is less biomass available to the fishery. Decreased gear efficiency may be significant in fisheries where effort is spatially concentrated and where derelict gear acts as an attractant to Derelict Fishing Gear Assessment Framework 12 target and bycatch species (e.g., as artificial habitat). If derelict gear attracts and/or captures a substantial amount of target or bycatch stocks, commercial and recreational fishers might catch and harvest less than they would otherwise (i.e., active gear will be less efficient in the presence of derelict gear). The economic costs of harvest reductions can be quantified in terms of lost revenues (commercial), lost utility (recreational), or increased fishing costs (commercial and recreational). Note that recreational fishers may derive non-consumptive benefits through catch-and-release fishing, in which case changes in catch, and the associated changes in utility, would be most relevant. To evaluate economic losses arising from harvest reductions, commercial and/or recreational harvests should be modeled. A common specification for harvest is: H  qEX 1. In equation (1), harvests (H) are specified as a function of effort (E), the stock of fish (X), and a scale factor known as a catchability coefficient (q). Derelict gear may affect both the stock of fish (stock reduction) as well as catchability (efficiency reduction). The total difference in harvest attributable to derelict gear can be specified as: 2. H  H  H  qEX  qEX where , q , and X are the harvest, catchability, and fish stock, respectively, in the absence of derelict gear. Equation (2) is thus the difference between harvests with and without derelict gear. If derelict gear significantly reduces the stock or gear efficiency, (2) would yield a negative value indicating harvests would be higher if derelict gear were not present. The data needed to evaluate (2) will depend on which mechanism of harvest loss is being modeled. In all cases, data on harvests, effort and stock is required. When modeling harvest losses arising from stock reductions, it is necessary to estimate X . This can be done using population models which explicitly incorporate the effects of derelict gear (e.g., as an additional source of mortality). Data on derelict gear mortality rates and other factors entering a model of stock dynamics (e.g., growth and maturity rates) would therefore be necessary in estimating X . Modeling losses due to decreased gear efficiency requires estimating q . An estimate of q can be derived through comparison of harvests across areas and/or times when derelict gear is and is not present, or is present to varying degrees, controlling for the effects of differences in effort and stock. Scheld et al.8 derived an estimate of q by incorporating derelict gear removals into a spatially explicit harvest model. Data on the amounts and location of derelict gear (or removals) and harvest is required to estimate q . 5.4 Habitat Damage Derelict gear can cause physical damage to coastal and marine habitats, primarily as a result of gear movement and scouring. The economic costs of habitat damage may be significant in areas with sensitive or ecologically important habitats, where strong currents, tides, or storm events cause substantial movement of derelict gear. Assessing the economic costs of habitat damage requires estimates of habitat value and the response of habitat to derelict gear. Habitat may be valued as an input to production of ecosystem goods and services (e.g., clean water or biomass) or as an offset to expenditures on man-made goods/services (e.g., for storm protection). Values for these goods and services can be drawn from available data and the Derelict Fishing Gear Assessment Framework 13 published literature or estimated by the analyst. Once value estimates are obtained, a mechanistic model is necessary to understand how a particular habitat provides the good or service and how changes in the amount or extent of that habitat affect provision, and therefore habitat value. The impact of derelict gear on a specific habitat can be found through observation and experimentation. Once habitat response to derelict gear has been determined, this information can be used together with value estimates and the mechanistic model to assess changes in value resulting from changes in the amount or location of derelict gear. Total costs or losses from derelict gear can be found through counterfactual comparisons (i.e., comparing habitat value with and without derelict gear). Data requirements will depend on the habitat being valued and the specific empirical models used, though in general, data on the amounts and values of ecosystem goods and services provided by the habitat, as well as habitat response to derelict gear will be necessary. 5.5 Navigational Hazard Derelict gear may pose a navigational hazard to commercial or recreational vessel traffic. Economic costs result from necessary changes in navigation to avoid derelict gear as well as damage to vessels and equipment. Costs may be significant in areas with heavy commercial or recreational traffic. Quantifying the costs of derelict gear resulting from increased hazards to navigation requires a model of route planning or navigation decision-making and/or vessel and equipment damage assessments. Fuel, labor, and materials/equipment costs of vessel traffic patterns with and without derelict gear can then be evaluated and used to assess costs. Data on vessel traffic, vessel and equipment damage, as well as amounts and location of derelict gear would be necessary to evaluate these costs. 5.6 Replacement Gear Fishers may ultimately need to replace the derelict gear. Replacement gear represents an economic loss if it would have been more cost effective had the gear not become derelict. These costs may be significant when gear is costly, when behavioral changes to reduce loss are inexpensive, or if derelict gear physically damages active gear. Note that simply summing expenditures on new gear to replace that which becomes derelict will likely produce an overestimate of economic losses. To accurately assess the economic costs of replacement gear, depreciation in value for the gear which became derelict would need to be estimated. Additionally, if gear loss could have been avoided through changes in fishing practices (e.g., spatial or temporal shifts in effort), the costs of such behavioral modifications should be deducted (i.e., current fishing practices offer cost savings which should be accounted for). If derelict gear physically damages active gear through movement or entanglement, the costs of these damages should also be included. The replacement costs or lost value of purposely abandoned gear should not be considered—had the gear been of any value to the fisher it would not have been abandoned. The data necessary to evaluate replacement gear costs include the cost of new gear, depreciation rates, age of gear when it becomes derelict, the costs of behavioral modifications to avoid loss, and estimates of damage to actively fished gear. Derelict Fishing Gear Assessment Framework 14 6. Detecting and Removing Derelict Fishing Gear 6.1 Overview Derelict fishing gear detection and removal methods depend upon the type of gear and the physical characteristics of the environment in which the lost gear is located. However, some basic guidance and essential baseline data collection is applicable across a suite of stationary gear types and can be helpful in providing critical information on the ecological and/or economic impacts resulting from derelict gear. 6.2 Derelict Gear Detection In remote sensing there are generally two important criteria: detection and identification. Detection refers to the observation of a potential object of interest while identification reveals the object as a specific type.55 Various remote sensing survey methods exist in regard to derelict fishing gear ranging from visual to acoustic (Table 1). Visual surveys can utilize divers with or without surface supplied air or involve surface observers scanning an area for easily observable lost gear. Acoustic surveys (i.e. side scan sonar) are a form of remote sensing where the acquisition of information is obtained using a detection system based on the reflection of underwater sound waves. Morison and Murphy56 summarize projects using these methods to detect and identify derelict traps. The basic approach and application of these methods could be used to detect and identify any type of derelict fishing gear. Side scan sonar has been shown to be very useful in detecting and identifying derelict fishing gear not only because of the detection capability but also the GPS tagging of the object location for future recovery. Bottom trawling or grappling is a direct survey method that usually involves dragging a hook and line type device on the bottom of an area in an effort to capture derelict gear. NRC57 describe both trawling and diver type surveys with accompanying pros and cons. The appropriate survey method used will vary depending on the survey area environment and gear target type and ultimately involves a trade-off between data acquisition and expense. Regardless of the survey method used, training on various images that will be potentially encountered during the survey is critical. The collection of an image database from which image calibration and validation can be conducted increases the ability of the observer to differentiate between derelict gear and other marine debris and is an important component of any survey.13 An image catalog can be developed by deploying target gear in various aspect configurations and degrees of decay from which observers can be trained on image identification. A vital piece of information to obtain during derelict gear surveys is the specific location of the identified item. In sonar surveys, this is usually obtained via GPS 12 or, in the case of towed divers, by vessel/diver coordinated timed transects.58 Collection of spatially explicit data increases the likelihood of derelict fishing gear recovery and provides enhanced analysis in regard to the geospatial distribution of derelict gear, relationships to harvest and fishing effort, and bycatch distribution as well as the capacity to resurvey the area to obtain gear loss rates. Derelict Fishing Gear Assessment Framework 15 Table 1. General survey methods to detect and remove derelict fishing traps and nets. Method Expertise Advantages Challenges References Side Scan Sonar – Hull Mounted Minimal Low cost, GPS marked gear, can be programmed to scan transects Limited to relatively shallow zones (15m). Limited utility in high relief areas 12, 59 Side Scan Sonar - AUV High GPS marked gear, can be programmed to scan transects, can scan deeper than hull mounted Limited utility in high relief areas, entanglement risk, expensive 60 Side Scan Sonar Towed Moderate GPS marked gear, can be programmed to scan transects, can scan deeper than hull mounted Limited utility in high relief areas, moderately expensive 13, 30 Diver Moderate (shallow) to High (deep) Maneuverability, can both identify and remove debris Survey area limited, water clarity limited, expensive depending on depth, entanglement risk 57, 61 Towed diver Moderate Versatile in various bottom structure conditions, potential for multiple data collection along survey transect Survey area limited, water clarity limited, restricted movement, some entanglement risk 58, 62, 63 Cameras – submersible /towed High GPS marked gear, visual record, access to depths Limited camera views, limited in low visibility situations, can be expensive depending on platform 60, 64 Bottom trawling / grappling by transect Minimal Removal of gear during survey Depending on transect length – limited geospatial information on gear, benthic habitat impacts due to indiscriminate bottom trawling / grappling 32 Visual observation at or near the surface Minimal Removal of abandoned gear with attached floats Limited to easily observable gear, misses submerged / un-buoyed gear 65 6.3 Derelict Gear Removal Determining the regional primary drivers for gear loss (i.e. accidental, abandonment, storms, vandalism, equipment failure) is critical for the implementation of any management action. Information on the primary drivers of gear loss can be gathered by conversations with fishers and management personnel or through surveys of the lost gear. Stakeholders, in particular commercial fishers and resource management Derelict Gear Removal Highlights personnel, have regional expertise on derelict gear location and the causes of gear loss  Engage commercial fishers and resource critical to a successful program. Early managers early in the planning involvement of commercial fishers and  Where appropriate, hire commercial fishers for data collection and derelict gear removal management personnel in the program  Obtain GPS coordinates for all gear removed design, data collection, and program  Take pictures of all gear removed modification provides a positive interactive  Collect data on derelict gear bycatch experience with the scientific and governmen Collect data on derelict gear condition 66 tal aspects of natural resource management. Derelict Fishing Gear Assessment Framework 16 Establishing two-way interactions that produce positive outcomes enhances program legitimacy and can be crucial to implementation of management scenarios while direct involvement in data collection enables peer to peer knowledge transfer among the stakeholder community. The North Carolina Coastal Federation, in partnership with NC Sea Grant, developed a simple electronic tablet app for derelict trap data collection (http://www.nccoast.org/crabpot) which they have successfully implemented with NC commercial fishers. A protocol for survey and removal of derelict Educational materials on derelict fishing gear trap removal and data collection are available at 1) Determine Primary Drivers of Gear Loss http://ccrm.vims.edu/marine_debris a) Survey of fishers _removal/. b) Survey of management personnel 2) Survey of Derelict Gear Retrieval methods for derelict gear a) Image catalog: Quality Assurance/Quality Control differ based upon the gear type and (QA/QC) for derelict gear identification the habitat in which the gear is 61,66 b) Development of a survey design located and, in some cases due to c) GPS location of derelict gear (critical information) the potential for damage to benthic i) Location relative to sensitive habitat habitats during removal efforts, it ii) Location relative to boating activity may be necessary to assess the cost iii) Location relative to fishery management areas / benefit of removing derelict gear from sensitive habitats. 1,4,12 iv) Spatial distribution and density 3) Removal of Derelict Gear a) Pictures of items collected and bycatch (also serves as QA/QC for removed gear) b) Bycatch Documentation i) Species ii) Sex iii) Size iv) Living/injured/dead c) Gear Documentation i) Functional/nonfunctional (still capable of trapping animals?) ii) Buoyed/un-buoyed iii) Location relative to sensitive habitat (seagrass, marsh, reefs, etc.) iv) Identifying features, labels, tags v) Escape panels present? (1) functional/nonfunctional vi) Biofouling on gear 4) Post Removal of Derelict Gear a) Match removal data with fishery management areas i) Harvest data ii) Fishing effort b) Dispose of gear responsibly Derelict Fishing Gear Assessment Framework 17 7. Evaluating Management Implications The framework presented in this document offers a strategy for characterizing derelict gear distributions and associated impacts to ecosystems and the fisheries they support. It provides guidance on the key data requirements, data collection methods, and analytic approaches that could be used to assess the scope and extent of the derelict gear problems within managed fisheries. Through modeling scenarios based on a recent Chesapeake Bay assessment,33 the framework identifies practical and effective management options that can be used for mitigating derelict gear impacts within a defined region. Furthermore, the framework can be used to structure existing and future derelict gear removal and reduction initiatives so that they collect and synthesize baseline data that is essential for informing management actions. Based on lessons learned from the Chesapeake Bay assessment, the following are key recommendations for assessing derelict gear distributions and mitigating associated environmental and economic impacts. 7.1 Spatial and Temporal Management of the Fishery Lesson learned: The way in which a fishery is managed can influence the amount of gear that becomes derelict and may ultimately determine the level to which derelict gear negatively impacts the fishery, target and non-target species, and habitats. Recommendation: Review spatial and temporal management of the fishery. Process: 1. Convene meetings of resource managers and fishers to discuss management and primary drivers of derelict gear within the fishery. 2. Identify regulations that may contribute to increased abundance of derelict fishing gear and their effects. This exercise could also help to guide the approach for identifying geographic locations where derelict gear may be more persistent than other areas. Examples of these types of regulations include: 2.1.Areas designated as no fishing for a particular gear or off limits to all fishing activity (e.g., marine protected areas) tend to have high densities of fishing gear along their boundaries, and may have greater concentrations of derelict fishing gear. 2.2.Temporal regulations that would increase the likelihood that certain gear types might become derelict. Gear loss could be attributed to instances when certain fishing areas are opened for short durations, or opened during seasons when weather events contribute to greater gear loss. 3. Evaluate how changes to regulations such as additional enforcement, modifying rules (e.g., allowing fishers to remove abandoned gear), spatial and temporal management changes, and other mitigation strategies or incentives (e.g., functional cull rings, biodegradable panels, derelict gear removal efforts) could modify the effects of derelict gear in a particular fishery or environment. 4. Develop scenarios and quantify how modifications identified above would change the effects of derelict gear in the fishery or environment. 5. Monitor to determine if any subsequently implemented actions are having the desired effect and recruit fishers in the data collection. Derelict Fishing Gear Assessment Framework 18 7.2 Data Reviews, Compilation, and Synthesis Lesson learned: The collection of spatially explicit data on derelict gear location, target and non target species harvest, and fishing effort allows for robust modeling of ecological and economic impacts for management consideration. Recommendation: Conduct comprehensive and exhaustive data reviews, compilation, and synthesis. Process: 1. Review active gear, derelict gear, and harvest data to determine spatial extent, collection method, and potential biases. 2. Review sensitive habitat spatial distribution data (e.g., locations of reefs, shellfish beds, aquatic vegetation). 3. Review spatial information on resource overlap between competing interests (e.g. commercial fishers, recreational boating). 4. Evaluate spatial resolution of data to determine appropriate alignment between management information (e.g., harvest and effort data) and location and abundance of derelict gear. 5. Modify collection and reporting of harvest and effort data to the appropriate spatial and temporal scales. 7.3 Data Collection Plans to Fill Critical Data Gaps Lesson learned: Critical information gaps on spatial patterns of human use hinder assessments of ecological and economic impacts from derelict gear. Recommendation: Develop and implement data collection plans to fill critical data gaps. Process: 1. Consider the use of participatory mapping when engaging managers and stakeholders to identify spatial use patterns. 2. Review and explore the use of secondary social science data to identify spatial demographic patterns over large areas and obtain economic profiles of users. 3. Develop and implement data collection plans such as user and fishing effort surveys to quantify recreational fishing effort and recreational uses. For example, anecdotal data and information from local fishers in the Chesapeake Bay suggest that recreational crabbing is a likely source of derelict gear, but the intensity of recreational fishing effort in the Chesapeake Bay is not fully known. Relating derelict gear location and removal activities to ecological and economic management implications, and understanding the factors affecting implementation of mitigating actions, are critical to addressing the global issue of derelict fishing gear. This Framework provides a tiered approach, based upon various levels of available data, to inform stakeholders and to help them adaptively manage derelict fishing gear impacts based upon the needs of the particular fishery. Derelict Fishing Gear Assessment Framework 19 Appendix A. Global Derelict Fishing Gear Estimates Table A-1. Documented effects on select trap fisheries of derelict fishing gear 5 constructed with wire or synthetic materials. (Modified from Bilkovic et al. ) Region (U.S.A. unless noted) Traps deployed Annual gear loss (%) Capture or mortality in derelict traps (individuals per trap per year unless noted) ~400,000 10 − 20% 21 (mortality) 13, 67 392,175 10 − 30% 50.6 (capture) 12, 66 68 References Blue Crab - Callinectes sapidus Maryland Virginia Virginia ~250,000 20% 18 (mortality) North Carolina >1 million 14 − 21% 12 − 17% 20-30 (mortality); 40.4 (capture) (44% mortality in derelict pots) Chesapeake Bay 600,000 – 800,000 12 - 20% 23 (mortality) totalling 3.3 million / year (~5% of harvest) 33 361,912 30 − 50% − 46 26,100 25% − 46 − 20 − 30% − 46 870,220 up to 100% 25.8 (mortality) 51,800 35 − 50% − 20 − 25% ≥ 33% mortality 70, 71 18 - 22% 3.0 – 6.8 20, 72 Florida - Gulf Coast Alabama Mississippi Louisiana Texas 29, 59, 69 10, 46 46 American Lobster - Homarus americanus Maine 3.2 million Caribbean spiny lobster - Panulirus argus Florida Keys ~500,000 Dungeness crab - Cancer magister British Columbia, Canada 26,250 11% 9.3 (7% of the catch in the Fraser River) 28 California - 100,000 traps - 73 Washington state − 12,193 traps 0.058 (mortality) (4.5% of recent harvest) 27 Alaska, southeast 23,000 92,000 - Capture: 4.47% of harvest Mortality: 2.2% of the harvest 30 − 10 – 20% 39% mortality 74 Tanner crab - Chionoecetes bairdi Bering Sea, Alaska Red king crab - Paralithodes camtschatica Alaska − 10% 2−7% mortality 75 Womens Bay, Kodiak Island, Alaska - - 40 – 56% mortality 76 ~9,000 70% 3−223 14 217,929 24% 0.87 77 − − 17,442 20%, 88% Blue swimmer crab - Portunus pelagicus Queensland, Australia Octopus - Octopus vulgaris northeast Atlantic Ocean, south coast of Portugal Sablefish - Anoplopoma fimbria British Columbia, Canada 15−30% of the commercial catch (1977–1983) 78 Various fish and invertebrates Sultanate of Oman (also used in Arab Persian Gulf states) Derelict Fishing Gear Assessment Framework 80.51 kg 79, 80 20 Region (U.S.A. unless noted) Traps deployed Annual gear loss (%) Capture or mortality in derelict traps (individuals per trap per year unless noted) 6,500 10 - 20% 22.9 kg References Fish traps St. Thomas & St. John, U.S. Virgin Islands Barbados, Caribbean Iskenderun Bay, Istanbul, Turkey 81 - 25% - 82, 83 ~5000 11% (traps) 13-65% (nets) - 84, 85 40,000 50% - 86 ~1 million 15.6% - 87 Lobster and fish traps Guadeloupe (French West Indies Eel traps Korea Table A-2. Gear loss and global landings for major crustacean pot and trap fisheries. 8 Annual averages, 2003-2012. (after Scheld et al. ) Species Blue swimmer crab Portunus pelagicus American lobster Homarus americanus Blue crab Callinectes sapidus Queen crab/snow crab Chionoecetes opilio Edible crab Cancer pagurus Dungeness crab Metacarcinus magister Spiny lobster Panulirus argus King crab Paralithodes camtschaticus Stone crab (claws only) Menippe mercenaria TOTAL Gear Loss 5 (%) 70 Landings (MT) 173,647 Revenues (US$) $199Ma 20-25 100,837 $948M Major Producers China, Philippines, Indonesia, Thailand, Vietnam Canada, USA 10-50 98,418 $152M USA NA 113,709 $401M NA 45,783 $49Mb 11 35,659 $169M Canada, St. Pierre and Miquelon (France), USA United Kingdom, Ireland, Norway, France USA, Canada 10-28 34,868 $500M90 10 10,137 $99M Bahamas, Brazil, Cuba, Nicaragua, Honduras, USA USA NA 2,502 $24M USA 615,560 $2.5B Notes on Table A-2: Data from: NOAA Office of Science and Technology, National Marine Fisheries Service, Commercial fisheries statistics http://www.st.nmfs.noaa.gov/st1/commercial; Food and Agriculture Organization, United Nations, Fisheries and Aquaculture Department, http://www.fao.org/fishery/search/en, Fisheries and Oceans Canada http://www.dfompo.gc.ca/stats/commercial/sea-maritimes-eng.htm. a 88 Based on an average price of US $1.15/kg. b Based on 2004-2012 average price of US $1.07/kg. 89 Derelict Fishing Gear Assessment Framework 21 References 1. Arthur, C., Sutton-Grier, A., Murphy, P., Bamford, H. 2014. Out of sight but not out of mind: Harmful effects of derelict traps in selected U.S. coastal waters. Marine Pollution Bulletin. 86: 19-28. 2. United States Congress (112th), 2012. Marine Debris Act Amendments of 2012 (H.R. 1171) and United States Congress (109th), 2006. Marine Debris Research, Prevention, and Reduction Act. S.362. (33 U.S.C. 1951-1958 (2006)) 3. Hérbert, M., G. Miron, M. Moriyasu, R. Vienneau, P. DeGrâce. 2001. Efficiency and ghost fishing of snow crab (Chionoecetes opilio) traps in the Gulf of St. Lawrence. Fisheries Research 52:143-153. 4. Matsuoka, T., Nakashima, T. & Nagasawa, N. 2005. A review of ghost fishing: scientific approaches to evaluation and solutions. Fish. Sci. 71:691–702. 5. Bilkovic, D., K. Havens, D. Stanhope, K. Angstadt. 2012. The use of fully biodegradable panels to reduce derelict pot threats to marine fauna. Conservation Biology 26 (6): 957-966. 6. Uhlmann, S. S. & Broadhurst, M. K. 2015. Mitigating unaccounted fishing mortality from gillnets and traps. Fish Fish. 16:183–229. 7. Macfadyen, G., Huntington, T. & Cappell, R. 2009. Abandoned, lost or otherwise discarded fishing gear (UNEP Regional Seas Reports and Studies, No. 185; FAO Fisheries and Aquaculture Tech. Paper, No. 523; Rome, 2009). 8. Scheld, A. M., Bilkovic, D, Havens, K.J., 2016. The dilemma of derelict Gear. Sci. Rep. 6:19671; doi: 10.1038/srep19671. 9. Smolowitz, R.J. 1978. Trap design and ghost fishing: an overview. Mar. Fish. Rev. 40:2-8. 10. Guillory, V. 1993. Ghost fishing by blue crab traps. North Am. J. Fish. Manag. 13: 459–466. 11. Brown, J. & Macfadyen, G. 2007. Ghost fishing in European waters: Impacts and management responses. Mar. Policy 31: 488–504. 12. Havens, K. J., Bilkovic, D. M., Stanhope, D. M., Angstadt, K. T. & Hershner, C. 2008. The effects of derelict blue crab traps on marine organisms in the lower York River, Virginia. North Am. J. Fish. Manag. 28:1194–1200. 13. Slacum H.W., Jr., S. Giordano, J. Lazar, D. Bruce, C. Little, D. Levin, H. J. Dew- Baxter, L. Methratta, D.Wong, R. Corbin. 2009. Quantifying the effects of derelict fishing gear in the Maryland portion of Chesapeake Bay. Prepared for the NOAA Marine Debris Program. July 2009. 14. Campbell, M.J., W.D. Sumpton. 2009. Ghost fishing in the pot fishery for blue swimmer crabs Portunus pelagicus in Queensland, Australia. Fisheries Research 95: 246-253. 15. Bae, B. S., An, H. C., Jeong, E. C., Park, H. H., Park, S. W., & Park, C. D. 2010. Fishing power estimation of biodegradable traps in the East Sea. Journal of the Korean society of Fisheries Technology, 46(4): 292-301. 16. Carr, H. A., and J. Harris. 1997. Ghost fishing gear: Have fishing practices during the past few years reduced the impact? Pages 141–151 in J. M. Coe and D. B. Rogers, editors. Marine debris, sources, impacts, and solutions. Springer, New York. 17. Slack-Smith, R. J. 2001. Fishing with traps and pots (No. 26). Food & Agriculture Org. Derelict Fishing Gear Assessment Framework 22 18. Kennedy, V. S., M. Oesterling, W. A. Van Engel. 2007. History of blue crab fisheries on the U.S. Atlantic and Gulf coasts. Pages 655–709 in V. Kennedy and L. Cronin, editors. The Blue crab: Callinectes sapidus. Maryland Sea Grant College, College Park. 19. Lewis, C. F., Slade, S. L., Maxwell, K. E. & Matthews, T. R. 2009. Lobster trap impact on coral reefs: Effects of wind‐driven trap movement. New Zealand Journal of Marine and Freshwater Research, 43 (1): 271-282. 20. Butler, C. B., & Matthews, T. R. 2015. Effects of ghost fishing lobster traps in the Florida Keys. ICES Journal of Marine Science: Journal du Conseil 72(suppl 1), i185-i198. 21. Parrish, F. A. & Kazama, T. K. 1992. Evaluation of ghost fishing in the Hawaiian lobster fishery. Fish. Bull. 90:720–725. 22. Godoy, H., Furevik, D. & Stiansen, S. 2003. Unaccounted mortality of Red King Crab (Paralithodes camtschaticus) in deliberately lost pots off Northern Norway. Fish. Res. 64: 171–177. 23. Sturdivant, S. K. & Clark, K. L. 2011. An evaluation of the effects of blue crab (Callinectes sapidus) behavior on the efficacy of crab pots as a tool for estimating population abundance. Fish. Bull. 109:48–55. 24. Newman, S. J., Skepper, C. L., Mitsopoulos, G. E., Wakefield, C. B., Meeuwig, J. J., & Harvey, E. S. 2011. Assessment of the potential impacts of trap usage and ghost fishing on the Northern Demersal Scalefish Fishery. Reviews in Fisheries Science, 19(2): 74-84. 25. Renchen, G.F., S.J. Pittman, R. Clark, C. Caldow, S. Gall, D. Olsen, R.L. Hill. 2014. Impact of derelict fish traps in Caribbean waters: an experimental approach. Bull. Mar. Sci. 90(2): 551563. 26. Eggleston, D., Etherington, L., & Elis, W. 1998. Organism response to habitat patchiness: species and habitat-dependent recruitment of decapod crustaceans. Journal of experimental marine biology and ecology, 223(1): 111-132. 27. Antonelis, K., Huppert, D., Velasquez, D. & June, J. 2011. Dungeness crab mortality due to lost traps and a cost-benefit analysis of trap removal in Washington State waters of the Salish Sea. North Am. J. Fish. Manag. 31:880–893. 28. Breen, P. A. 1987. Mortality of Dungeness crabs caused by lost traps in the Fraser river estuary, British Columbia. North American Journal of Fisheries Management 7: 429–435. 29. Lee, D. 2009. North Carolina Derelict Crab Pots 101: shallow water efforts. Pages 28–34 in Morison and Murphy.56 30. Maselko, J., Bishop, G., & Murphy, P. 2013. Ghost fishing in the Southeast Alaska commercial Dungeness crab fishery. North American Journal of Fisheries Management, 33(2): 422-431. 31. Donaldson, A., Gabriel, C., Harvey, B. J., Carolsfeld, J., & Department of Fisheries and Oceans, Ottawa, ON (Canada); Canadian Science Advisory Secretariat, Ottawa, ON (Canada). 2010. Impacts of fishing gears other than bottom trawls, dredges, gillnets and longlines on aquatic biodiversity and vulnerable marine ecosystems. Canadian Science Advisory Secretariat= Secrétariat canadien de consultation scientifique. 32. Cho, D. 2011. Removing derelict gear from the deep seabed of the East Sea. Marine Policy 35: 610-614. 33. Bilkovic, D.M., Slacum, Jr., H.W., Havens, K.J., Zaveta, D., Jeffrey, C.F.G., Scheld, A.M., Stanhope, D., Angstadt, K., Evans, J.D. 2016. Ecological and Economic Effects of Derelict Fishing Gear in the Chesapeake Bay: 2015/2016 Final Assessment Report. Prepared for Marine Debris Program, Office of Response and Restoration, National Oceanic and Atmospheric Administration. Derelict Fishing Gear Assessment Framework 23 34. Miller, T.J., Wilbert, M.J., Colton, A.R., Lipcius, R.N., Ralph, G.M., Johnson, E.G., and Kaufman, A.G. 2011. Stock assessment of blue crab in Chesapeake Bay. Final report to NOAA Chesapeake Bay Office, Annapolis, MD. 35. Van Engel, W.A., 1958. The blue crab and its fishery in Chesapeake Bay: Part 1 –Reproduction, early development, growth, and migration. Commercial Fisheries Review 20 (6):6–17. 36. Turner, H.V., Wolcott, D.L., Wolcott, T.G., Hines, A.H. 2003. Post-mating behavior, intramolt growth, and onset of migration to Chesapeake Bay spawning grounds by adult female blue crabs, Callinectes sapidus Rathbun. Journal of Experimental Marine Biology and Ecology 295: 107–130. 37. Aguilar, R., Hines, A.H., Wolcott, T.G., Wolcott, D.L., Kramer, M.A., Lipcius, R.N. 2005. The timing and route of movement and migration of postcopulatory female blue crabs, Callinectes sapidus Rathbun, from the upper Chesapeake Bay. Journal of Experimental Marine Biology and Ecology 319: 117–128. 38. Jensen, O.P., Seppelt, R., Miller, T.J., Bauer, L.J. 2005. Winter distribution of blue crab Callinectes sapidus in Chesapeake Bay: application and cross-validation of a two stage generalized additive model. Marine Ecology Progress Series 299:239–255. 39. Jensen, O.P., Miller, T.J. 2005. Geostatistical analysis of the abundance and winter distribution patterns of the blue crab Callinectes sapidus in Chesapeake Bay. Transactions of the American Fisheries Society 134:1582–1598. 40. Smith, S.G., Chang, E.S. 2007. Molting and growth. In: Kennedy, V.S., Cronin, L.E. (Eds.), The Blue Crab: Callinectes sapidus. University of Maryland Sea Grant College, College Park, Maryland, pp. 197–254. 41. Slacum, Jr., H. W., E. Methratta, J. Dew-Baxter, D. Wong, and R. Corbin. 2011. 2010 Monthly commercial blue crab effort survey in the Virginia portion of Chesapeake Bay. Annual report to Virginia Marine Resource Commission. 42. Slacum, Jr., H. W., E. Methratta, J. Dew-Baxter, D. Wong, and R. Corbin. 2012. 2011 Monthly commercial blue crab effort survey in the Maryland portion of Chesapeake Bay. Annual report to Maryland Department of Natural Resources. 43. Caldow, C, Monaco, M. E., Pittman, S.J., Kendall, M. S., Goedeke, T. L., Menza, C., Kinlan, B. P., Costa, B. M. 2015. Biogeographic assessments: a framework for information synthesis in marine spatial planning. Marine Policy 51: 423-432. 44. Müller, F., Breckling, B., Jopp, F., Reuter, H. 2011. What are the general conditions under which ecological models can be applied? In: Jopp, F., Reuter, H., Breckling, B. (Eds.), Modelling Complex Ecological Dynamics. Springer, Berlin, Heidelberg, pp. 13–28. 45. Fath, B., Patten, B.C. 2000. Goal Functions and Network Theory. Presented in Porto Venere May 2000 at the Second Conference on Energy. 46. Guillory, V., McMillen-Jackson, A., Hartman, L., Perry, H., Floyd, T., Wagner, T., and Graham, G. 2001. Blue crab derelict raps and trap removal programs. Gulf States Marine Fisheries Commission, Ocean Springs, Mississippi. 47. Mitro, M. G. 2003. Demography and viability analysis of a diamondback terrapin population. Canadian Journal of Zoology 81:716–726. 48. Hart, K. M., and L. B. Crowder. 2011. Mitigating by-catch of diamondback terrapins in crab pots. Journal of Wildlife Management 75:264–272. 49. Gilman, E., Suuronen, P., Hall, M. and Kennelly, S. 2013. Causes and methods to estimate cryptic sources of fishing mortality. Journal of Fish Biology, 83(4):766-803. Derelict Fishing Gear Assessment Framework 24 50. Yeon, I., Song, M.Y., Shon, M.H., Hwang, H.J. and Im, Y.J.2010. Possible new management measures for stock rebuilding of blue crab, Portunus trituberculatus (Miers), in western Korean waters. Proceedings of Korean Applied Industrial Sciences, 5(2): 35. 51. Breen, M., Graham, N., Pol, M., He, P., Reid, D. and Suuronen, P. 2016. Selective fishing and balanced harvesting. Fisheries Research doi:10.1016/j.fishres.2016.03.014. 52. NOAA Marine Debris Program. 2015. 2015 Report on the impacts of “ghost fishing” via derelict fishing gear. Silver Spring, MD. 25 pp. 53. Gilardi, K.V.K, Carlson-Bremer, D., June, J.A., Antonelis, K. 2010. Marine species mortality in derelict fishing nets in Puget Sound, WA and the cost/benefits of derelict net removal. Marine Pollution Bulletin 60: 376-382. 54. FAO. 2016. Abandoned, lost or otherwise discarded gillnets and trammel nets: methods to estimate ghost fishing mortality, and the status of regional monitoring and management, by E. Gilman, F. Chopin, P. Suuronen and B. Kuemlangan. FAO Fisheries and Aquaculture Technical Paper No. 600. Rome. Italy 55. Havens, K.J. and Sharp, E.J. 2015. Thermal Imaging Techniques to Survey and Monitor Animals in the Wild: A Methodology. Academic Press, 354 pp. 56. Morison, S., and Murphy, P. (eds). 2009. Proceedings of the NOAA submerged derelict trap methodology detection workshop. Technical memorandum NOS-OR&R-32. National Oceanic and Atmospheric Administration, Silver Spring, Maryland. 57. NRC (Natural Resources Consultants, Inc.). 2013. Deepwater derelict fishing gear removal protocols. Report prepared for the National Oceanic and Atmospheric Administration and the Northwest Straits Marine Conservation Foundation. Natural Resources Consultants, Inc. Seattle, WA, 25p. 58. Uhrin, A.V., Matthews, T.R., and Lewis, C. 2014. "Lobster trap debris in the Florida Keys National Marine Sanctuary: distribution, abundance, density, and patterns of accumulation." Marine and Coastal Fisheries 6.1: 20-32. 59. Voss, C.M., J.A. Browder, A. Wood, A. Michaelis. 2015. Factors driving the density of derelict crab pots and their associated bycatch in North Carolina waters. Fishery Bulletin 113:378390. 60. Olsen, D., & Hill, R. L. 2012. St. Thomas and St. John Trap Fishery: Present Status and Trap Loss. Survey and Impact Assessment of Derelict Fish Traps in St. Thomas and St. John, US Virgin Islands, 5. 61. Good, T. P., June, J. A., Etnier, M. A., & Broadhurst, G. 2010. Derelict fishing nets in Puget Sound and the Northwest Straits: Patterns and threats to marine fauna. Marine Pollution Bulletin, 60(1): 39-50. 62. Donohue, M. J., Boland, R. C., Sramek, C. M., & Antonelis, G. A. 2001. Derelict fishing gear in the Northwestern Hawaiian Islands: diving surveys and debris removal in 1999 confirm threat to coral reef ecosystems. Marine Pollution Bulletin, 42(12): 1301-1312. 63. Martens, J. and Huntington, B.E. 2012. Creating a GIS-based model of marine debris “hot spots” to improve efficiency of a lobster trap debris removal program. Marine pollution bulletin, 64(5): 949-955. 64. Watters, D. L., Yoklavich, M. M., Love, M. S., & Schroeder, D. M. 2010. Assessing marine debris in deep seafloor habitats off California. Marine Pollution Bulletin, 60(1): 131-138. 65. Anderson, J.A. and Alford, A.B. 2014. Ghost fishing activity in derelict blue crab traps in Louisiana. Marine Pollution Bulletin, 79(1): .261-267. Derelict Fishing Gear Assessment Framework 25 66. Havens, K., Bilkovic, D.M., Stanhope, D. and Angstadt, K. 2011. Fishery failure, unemployed commercial fishers, and lost blue crab pots: an unexpected success story. Environmental Science & Policy, 14(4): 445-450. 67. Giordano, S., Lazar, J., Bruce, D., Little, C., Levin, D., Slacum Jr., H.W., Dew-Baxter, J., Methratta, L., Wong, D. and Corbin, R. 2011. Quantifying the impacts of derelict blue crab traps in Chesapeake Bay. In: Carswell, B., McElwee, K. and Morison, S., eds. Proceedings of the Fifth International Marine Debris Conference. Technical Memorandum NOS OR&R-38. NOAA: Silver Spring, MD. 68. Bilkovic, D., K. Havens, D. Stanhope, K. Angstadt. 2014. Derelict fishing gear in Chesapeake Bay, Virginia: spatial patterns and implications for marine fauna. Marine Pollution Bulletin 80: 114-123. 69. NCDMF. 2013. North Carolina Blue Crab (Callinectes sapidus) Fishery Management Plan: Amendment 2. North Carolina Department of Environment and Natural Resources. North Carolina Division of Marine Fisheries. Morehead City, NC. 408 pp. 70. Smith, M., L. Ludwig, and C. Wilson. 2011. Investigating the ‘ghost fishing’ capacity of derelict lobster traps. Pages 331–333 in B. Carswell, K. McElwee, and S. Morison, editors. Proceedings of the Fifth international marine debris conference. Technical memorandum NOS-OR&R-38. National Oceanic and Atmospheric Administration, Silver Spring, Maryland. 71. Sheldon, W. W., and R. L. Dow. 1975. Trap contributions to losses in the American lobster fishery. U.S. National Marine Fisheries Service Fishery Bulletin 73:449–451. 72. Matthews, T. R., and Uhrin, A. V. 2009. Lobster trap loss, ghost fishing, and impact on natural resources in Florida Keys National Marine Sanctuary. In Morison and Murphy.56 73. Kennedy, R. 1986. The Dungeness crab of North America. Safish (South Australian Department of Fisheries) 9:35. 74. Kimker, A. 1994. Tanner crab survival in closed pots. Alaska Fishery Research Bulletin 1:179–183. 75. High, W., and D. Worlund. 1979. Escape of king crab (Paralithodes camtschatica) from derelict pots. Technical report NMFS SSRF-734. National Oceanic and Atmospheric Administration, Washington, D.C. 76. Long, W.C., Cummiskey, P.A. and Munk, J.E. 2014. Effects of ghost fishing on the population of red king crab (Paralithodes camtschaticus) in Womens Bay, Kodiak Island, Alaska. Fishery Bulletin, 112(2-3):101-112. 77. Erzini, K., L. Bentes, R. Coelho, P. G. Lino, P. Monteiro, J. Ribeiro, and J. M. S. Goncalves. 2008. Catches in ghost-fishing octopus and fish traps in the northeastern Atlantic Ocean (Algarve Portugal). Fisheries Bulletin 106: 321–327. 78. Scarsbrook, J. R., G. A. McFarlane, and W. Shaw. 1988. Effectiveness of experimental escape mechanisms in sablefish traps. North American Journal of Fisheries Management 8:158–161. 79. Al-Masroori, H., H. Al-oufi, J. L. McIlwain, and E. McLean. 2004. Catches of lost fish traps (ghost fishing) from fishing grounds near Muscat, Sultanate of Oman. Fisheries Research 69:407–414. 80. Al-Masroori, H., H. Al-oufi, and P. McShane. 2009. Causes and mitigations on trap ghost fishing in Oman: scientific approach to local fishers’ perception. Journal of Fisheries and Aquatic Science 4:129–135. 81. Clark, R., S.J. Pittman, T.A. Battista, and C. Caldow (eds.). 2012. Survey and impact assessment of derelict fish traps in St. Thomas and St. John, U.S. Virgin Islands. NOAA Technical Memorandum NOS NCCOS 147. Silver Spring, MD. 51 pp. Derelict Fishing Gear Assessment Framework 26 82. Wilson, S.B. 1983. A report on the trap fishing industry of Barbados. Division of Fisheries, Ministry of Agriculture, Food and Consumer Affairs, Barbados, 32pp. 83. Selliah, N. M. 2000. The Barbados trap fishery: Selecting biodegradable fasteners, testing effects of new gear regulations on catch rates, and determining current status. M.Sc. thesis, University of the West Indies, 82pp. 84. Özyurt, C. E.; Akamca, E.; Kiyağa, V. B.; Taşlıel, A. S. 2008. The rates and reasons of pots lost during a fishing season in Iskenderun Bay. Ege J. Fish. Aqua. Sci. 25(2): 147–151. 85. Yıldız, T. and Karakulak, F.S. 2016. Types and extent of fishing gear losses and their causes in the artisanal fisheries of Istanbul, Turkey. Journal of Applied Ichthyology, 32(3): 432-438. 86. Bouchereau, J.L. and Wuttichaï, B. 2010. October. Relationship between the wreck of smallscale fishing equipment and ghost fishing in Guadeloupe (French West Indies) with palliative proposals for a better long-lasting management of biodiversity. In Coastal Zone Asia-Pacific Conference-World Small-Scale Fisheries Congress. 87. Kim, S., Park, S.W. and Lee, K. 2014. Fishing performance of environmentally friendly tubular pots made of biodegradable resin (PBS/PBAT) for catching the conger eel Conger myriaster. Fisheries Science, 80(5): 887-895. 88. The World Bank. 2012. Evaluation of new fishery performance indicators (FPIs): a case study of the blue swimming crab fisheries in Indonesia and Philippines. Agriculture and Rural Development Discussion Paper 52. 89. Norwegian Directorate of Fisheries. 2015. Economic and biological figures for Norwegian fisheries, 2014. Fiskeridirektoratet, Bergen. 90. Gongora, M. 2010. Assessment of Spiny Lobster (Panulirus argus) of Belize based on fishery dependent data. United Nations University Fisheries Training Programme, Iceland. Acknowledgments We are grateful to the NOAA Marine Debris Program for their oversight and support to this work, performed under NOAA SciTech contract # DG133E-10-CQ-0034, Task Order 007. Derelict Fishing Gear Assessment Framework 27