• Interdisciplinary assessment of the skate fishery in the Gulf of Alaska

      Farrugia, Thomas J.; Seitz, Andrew C.; Kruse, Gordon H.; Criddle, Keith R.; Goldman, Kenneth J.; Tribuzio, Cindy A. (2017-12)
      Skates are common bottom-dwelling fishes and valuable non-target species in Gulf of Alaska fisheries. Although there is little demand for skates in the United States, markets in Europe and Asia are fueling desires for additional fishing opportunities on skates in Alaska. Management agencies, however, have been hesitant to allow increased harvests due to the lack of information on the ecology and population dynamics of skates, and the bioeconomics of skate fisheries. Specifically focusing on the two most commonly landedskate species in the Gulf of Alaska (GOA), the big skate (Beringraja binoculata) and the longnose skate (Raja rhina), I conducted an interdisciplinary project to address these knowledge gaps. First, I advanced our understanding of the movement patterns and habitat use of skates by satellite tagging big skates in the GOA. The results show that big skates can, and likely frequently do, travel long distances, cross management boundaries within the GOA, and spend more time in deeper waters than previously thought. Second, I used the insights from the movement study to develop the first stock assessment models for skates in the GOA. This represents an important improvement in modeling, laying the groundwork for the North Pacific Fishery Management Council to move from Tier 5 (more data limited) to Tier 3 (less data limited) harvest control rules, which should lead to increased confidence with which the total allowable catch (TAC) for skates is set. Finally, I used the sustainable harvest estimates from the stock assessment models to develop a model that examined the impacts of management decisions on the profitability of skate fishing. My research provides essential information about these understudied fishes, helping to improve the sustainability and profitability of skate harvests. Incorporation of best available science regarding skate ecology, population dynamics, and bioeconomics into fishery management fosters more responsible development of skate fisheries, sustainable fishery revenues, and employment, and reduces the risk of overfishing, stock collapse, and prolonged fishery closures. It is my hope that fishery management agencies and the fishing industry make use of the new information and insights presented in this dissertation to work collaboratively towards the responsible development of skate fisheries.
    • The reproductive biology and management of walleye pollock (Gadus chalcogrammus) in the Gulf of Alaska

      Williams, Benjamin C.; Kruse, Gordon; Criddle, Keith; Dorn, Martin; Quinn, Terrance II (2018-08)
      Ecosystem-based fishery management (EBFM) entails treating resource allocation and management as elements of a comprehensive framework that accounts for ecological linkages. The goal of EBFM is to maintain ecosystem resiliency in a manner that provides for the services desired e.g., fishery catch, species abundance, economic viability. Historically fisheries have been managed on a per species basis with a general focus on increasing or decreasing harvest rates. This management strategy often excludes meaningful processes such as interactions with other species, environmental changes, and economic effects of management changes. One feasible path for implementation of EBFM is through enhancement of existing single-species fishery management models. Contemporary age-structured stock assessment models generally use an estimate of spawning stock biomass (SSB), i.e., the biomass of female spawning fish, to approximate stock reproductive potential (RP). This approximation inherently assumes a proportional relationship between SSB and RP. Maturity at age or at length is a key aspect of reproductive biology that is central to estimating both RP and SSB. As a sequential augmentation to a single species management model the relationships among body condition, population abundance, the probability of being mature, relative fecundity, and environmental correlates were examined for female walleye pollock Gadus chalcogrammus in the Gulf of Alaska. Maturity data were corrected for spatial sampling bias using a mixed-effects generalized additive model. Once corrected for spatial bias, relationships between maturity, ocean temperature, body condition, ocean productivity (in the form of chlrophyll-a), and population abundance were explored. Estimates of fecundity were updated through the processing of archived samples and were also examined with mixed-effects generalized additive models to explore relationships between the previously listed covariates. Multiple measures of RP were examined to explore differences between methods currently incorporated into the stock assessment and updated measures of total egg production and time varying maturity. Walleye pollock body condition is density-dependent, declining with population abundance. However, after accounting for the effects of length, age, location, year, chlorophyll-a concentrations, summer ocean temperature and sample haul, condition has a positive effect on the probability of a fish being mature. Similarly, condition has a positive effect on relative fecundity, after accounting for length, age, egg diameter, chlorophyll-a concentrations, winter ocean temperature and sample haul. A positive relationship is observed between depth-integrated summer ocean temperature and maturity and depth-integrated winter ocean temperature and fecundity. Chlorophyll-a concentrations have a dome shaped relationship with maturity, peaking at 2.3 mg/m⁻³, and a negative relationship with fecundity. Variations in body condition have a direct influence on the estimated RP of the fish stock through both differences in the maturation schedule and total egg production. Over some periods these updated estimates of RP differ from estimates of female SSB from the annual stock assessment. Alternative estimates of annual RP, particularly total egg production, may provide better estimates of annual reproductive output than spawning stock biomass. In addition, relationships to density-dependent and density-independent factors provide informative predictions that can be incorporated into stock assessment analyses. Inclusion of spatially explicit information for walleye pollock maturity has implications for understanding stock reproductive biology and thus the setting of sustainable harvest rates used to manage this valuable fishery. Additionally, because management decisions have economic as well as biological consequences a suite of management strategies were simulated to examine the economic viability of a proposed small-vessel walleye pollock fishery in Alaska state waters in the Gulf of Alaska. As a case-study for straddling stocks, an agent-based model was developed to examine a suite of available federal and state management strategies as they relate to the economic viability of a nascent Alaska state-waters trawl fishery for walleye pollock that may develop after a long history of parallel state and federal waters management. Results of alternative strategies were compared in terms of indicators, such as variance of catch and quasi-rent value. Given the input characteristics of these simulations, the management strategy that produces the best overall improvements relative to status quo involved a federal-waters management strategy that allows for community-based cooperatives and an open access strategy in state-waters. Agent-based models may be used to inform managers of the underlying dynamics of catches and revenues in order to avoid unintended consequences of management decisions and to improve the likelihood of attaining fishery management objectives. This dissertation provides incremental additions to our knowledge of walleye pollock reproductive biology its spatial and temporal dynamics, and environmental correlates that may serve as ecological indices. These indices, coupled with an improved understanding of the socio-economic examinations of fishery management changes through agent-based modeling, may assist in producing more holistic management strategies, such as EBFM.