Traditional Stock Assessment Research Articles

Population estimation of bocaccio (Sebastes paucispinis) based on larval production

We estimate the stock size of bocaccio ( Sebastes paucispinis ) in the Southern California Bight based on the production of larvae. Area-weighted total length compositions of preflexion larvae in 2002 and 2003 were determined from ichthyoplankton survey data. Larval length-to-age transition matrices were estimated from data derived from the enumeration of daily otolith increments. In combination, these results were used to estimate daily rates of larval production and mortality during surveys conducted in those years. Daily rates of larval production were expanded to annual production rates using information developed from the long-term seasonal distribution of larvae in CalCOFI surveys. Total annual larval production was translated to total spawning biomass by calculation of the female population weight-specific fecundity derived from a length-based life table analysis of adult reproductive parameters. Our results indicate that there were 6953 – 10 656 metric tons of bocaccio biomass (males and females >16 cm fork length) in the Southern California Bight during the survey years, which agrees with biomass estimates from a traditional stock assessment. Unique and problematic aspects of the reproductive biology of Sebastes are discussed, including multiple spawning, weight-specific fecundity that depends on size, and bias in fecundity data gathered from vitellogenic ovaries.

A Management Strategy for Sedentary Nearshore Species that Uses Marine Protected Areas as a Reference

AbstractClassical approaches to fisheries stock assessment rely on methods that are not conducive to managing data‐poor stocks. Moreover, many nearshore rocky reef species exhibit spatial variation in harvest pressure and demographic rates, further limiting traditional stock assessment approaches. Novel management strategies to overcome data limitations and account for spatial variability are needed. With the ever‐increasing implementation of no‐take marine protected areas (MPAs), there is great potential for improving decision making in management through comparisons of fished populations with populations in MPAs at spatially explicit scales. We developed a management strategy that uses a combination of data‐based indicators sampled inside and outside of MPAs as well as model‐based reference points for data‐poor, sedentary nearshore species. We performed a management strategy evaluation of this MPA‐based decision tree model for a hypothetical population of grass rockfish Sebastes rastrelliger in California. We introduced process, observation, and model uncertainty in numerous scenarios and compared these scenarios with the precautionary approach currently used to manage data‐poor species. Our model consistently improved total catches while maintaining the biomass and spawning potential ratio at levels well within acceptable thresholds of management. We suggest further exploration of this MPA‐based management approach, and we outline a collaborative research program in the California Channel Islands that may well be suited for testing an experimental management procedure.

Assessing the stocks of the primary snappers caught in Northeastern Brazilian reef systems. 1: Traditional modelling approaches

Traditional stock assessment models were used to describe the current status of Lutjanus analis, L. chrysurus, L. jocu, L. synagris, and L. vivanus stocks off the coast of Northeast Brazil. The yield per recruit model, cohort analysis models (length-based cohort analysis (LCA), pseudo-cohort virtual population analysis (VPA), and real cohort VPA), and Thompson and Bell predictive model were used. The results showed LCA accumulates biases from pseudo-cohort VPA along with a high degree of sensitivity of growth parameters, whose determination is problematic and may not be adequate for many other reef species. Although the real cohort VPA based on short time series may also be limited, this method was determined to be the most appropriate within the traditional methods applied for the assessment of reef fish. However, the VPA model heavily depends on the accuracy of landing data and discards estimation and estimated natural mortality ( M), so caution should be applied when using such models. The conservative fishing mortality index F 0.1 was found to be the most adapted index when the uncertainty of parameter estimation and model limitations were considered. Consequently, a drastic reduction of 80–90% of the current fishing effort is recommended. Despite the VPA and Thompson and Bell model limitations, these models can provide insight into species/fishery dynamics and can also be utilised to analyse the effects of changes in fishing level on catches. In addition, the trends presented in this study are confirmed in a companion paper using more complex models.

Performance of artificial neural networks and discriminant analysis in predicting fishing tactics from multispecific fisheries

In the Mediterranean, bottom trawlers are multispecific and frequently apply different fishing tactics (FTs) even during the same fishing trip. Up to four individual FTs were distinguished in the study area where fishermen usually use mixtures of different FTs in daily fishing trips. Identifying the FTs actually performed is a key issue in traditional stock assessment methods. In this paper, we compare the performance of discriminant analysis and artificial neural networks for predicting FTs from the species composition of daily sale bills. We used data on the landings of each vessel from daily sale bills along with information on the FT actually performed, which was obtained by onboard observers who interviewed skippers about the FTs that they planned to employ. Discriminant analysis and artificial neural networks achieved comparable overall results and the success of predictions depended on both the sample size of the different data subsets (balancing) and the similarity between the species composition of different FTs (overlapping). Although the percentage of correct predictions was high for FTs with more than 25 cases, success decreased when the sample sizes were small. In addition, success in predicting mixtures of two different FTs increased with increasing dissimilarity between their corresponding species compositions.

Acoustic data from platforms of opportunity as fuel for ecosystem models

Echo sounders are a widely used tool for observing marine ecosystems. Traditionally, rigid designed surveys are used to integrate biomass to a global estimate, which is coupled to traditional stock assessment models. More recently, the focus has shifted from single species assessment to an ecosystem approach, taking into account ecosystem function and dynamics. As a response, several ecosystem models have been developed. The data requirements for these models are different than for the traditional models, and new sensors and platforms have been developed as a response to this need. However, examples on how to integrate these observations with these new models are scarce. We propose a state-space observation-space approach, and compare the results from a vertically structured eco-system model with observations from an echo sounder positioned at the Mid Atlantic ridge. We argue that this approach is less vulnerable to assumptions in the data analysis process, and discuss adequate observations regimes for these models.

Integrating catch-at-age and multiyear tagging data: a combined Brownie and Petersen estimation approach in a fishery context

A comprehensive framework for modelling data from multiyear tagging experiments in a fishery context is presented that incorporates catch data into the traditional Brownie tag–recapture model. Incorporation of catch data not only allows for improved estimation of natural and fishing mortality rates, but also for direct estimation of population size at the time of tagging. These are the primary quantities required to be estimated in stock assessments — having an approach for directly estimating them that does not require catch rates provides a potentially powerful alternative for augmenting traditional stock assessment methods. Simulations are used to demonstrate the value of directly incorporating catch data in the model. Results from the range of scenarios considered suggest that in addition to providing a precise estimate of population size (coefficients of variation ranging from ~15% to 30%), including catch data can decrease biases in the mortality rate estimates (natural mortality especially) and improve precision of fishing mortality rate estimates (by as much as 60% at age 1). The model is applied to southern bluefin tuna (Thunnus maccoyii) tag–recapture and catch data collected in the 1990s to provide estimates of natural mortality, fishing mortality, and abundance for five cohorts of fish.

Role of disease in abundance of a Pacific herring (Clupea pallasi) population

Disease significantly affects population abundance of Pacific herring (Clupea pallasi). Comprehensive epidemiological study of the Pacific herring population of Prince William Sound, Alaska, U.S.A., from 1994 to 2000 included complete necropsy examination of 230–500 fish each spring and 40–160 fish each fall (total n = 2983 fish). Mortality is best estimated, through modifications of an age-structured assessment model, using a disease index that combines the prevalence of viral hemorrhagic septicemia virus (VHSV) with the prevalence of ulcers. Risk factors for an epidemic include poor body condition and abundant recruitment before spawning in the spring. Prevalence of the pathogen Ichthyophonus hoferi increased as fish aged, but changes in I. hoferi prevalence were not related to changes in population abundance. Disease that caused an epidemic in 1998 (VHSV and ulcers) nearly disappeared from the population when changes in abundance were detected by traditional stock assessment methods in 1999. Disease significantly affects recruitment — the two lowest recruitment estimates on record, in 1994 and 1999, followed increased natural mortality of adults in 1993 and 1998.

A meta-assessment for elasmobranchs based on dietary data and Bayesian networks

We developed a new approach, meta-assessment, as a tool for identifying declining (and potentially threatened) fish stocks in situations where a lack of data precludes traditional stock assessments. Meta-assessments are models enhanced by the incorporation of other stock assessment results. We used this idea to estimate historic biomass trends for demersal elasmobranchs of the Irish Sea. Bayesian networks, constructed from published dietary data and resembling food webs, allowed us to incorporate into our estimates the results from virtual population analysis (VPA) for Irish Sea cod, sole, plaice and whiting. To assess accuracy, we used cross-validation, estimating historic biomass trends in each individual VPA species from trends in the other three plus trends in fishing effort. We compared predicted annual trends to those derived from VPA and found 66% accuracy. We also compared biomass trends estimated from annual trawl surveys to corresponding network predictions, recovering survey trends correctly 61% of the time for elasmobranchs, 78% of the time for gurnards (Triglidae) and 89% for bib and pout (Trisopterus spp.). Results suggest that of the 11 elasmobranchs examined, the angel shark (Squatina squatina) increased in biomass least frequently from 1987 to 1997, a view consistent with survey results. Our approach also suggested a marked decline in common skate (Dipturus batis) over the period 1965–1978, during which time the skate was extirpated from the Irish Sea. We conclude that meta-assessment can serve as an exploratory method for identifying potentially threatened stocks, where even landings data are unavailable.

A General Framework for Developing Sequential Fisheries Models

A sequential fisheries model relates observed data to the biological dynamics of an underlying stock. Either model component, dynamic or observational, can be subject to statistical variation. Current fisheries literature includes models with (1) variable dynamics and no observation error, (2) deterministic dynamics and observations subject to measurement error, and (3) combined dynamic and measurement variability. This paper presents a general framework for developing sequential fishery models and estimating model parameters from available data. The framework encompasses most traditional stock assessment models and suggests new, potentially useful extensions. It generalizes the conventional definition of state space model to include nonlinear equations with nonnormal error. The paper rigorously compares two paradigms (KF: Kalman filter, EV: errors in variables) used for parameter estimation. Each paradigm is formulated in both frequentist and Bayes contexts, where Bayes is shown to be most appropriate for the EV paradigm. Model design concepts are illustrated with a simple example oriented to catch data and a more complex example with catch-at-age data. Because the framework forces essential questions to be asked about underlying processes, observed data, and sources of variability, it can help delineate the limits of current knowledge and establish rational priorities for future data collection.