A simple implementation of the statistical modelling framework Gadget for cod in Icelandic waters

Taylor, Lorna; Begley, James; Kupca, Vojtech; Stefansson, Gunnar

doi:10.2989/ajms.2007.29.2.7.190

Cited by 23 publications

(29 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Temperature-dependent predation, foraging, metabolic, and growth rates are common in more complex spatially-explicit food web or whole of ecosystem models such as GADGET (e.g., Howell and Bogstad 2010;Taylor et al, 2007), Atlantis (e.g., Fulton et al, 2011;Kaplan et al, 2012;, and FEAST (Ortiz et al, this issue). Temperature functions for growth and predation can also be incorporated into MSCAA models, allowing this class of models to be used to evaluate interacting climate, trophodynamic, and fishery influences on recommended fishing mortality rates.…”

Section: )mentioning

confidence: 99%

A comparison of fisheries biological reference points estimated from temperature-specific multi-species and single-species climate-enhanced stock assessment models

Holsman

Ianelli

Aydin

et al. 2016

Deep Sea Research Part II: Topical Studies in Oceanography

View full text Add to dashboard Cite

A comparison of fisheries biological reference points estimated from temperature-specific multi-species and single-species climate-enhanced stock assessment models, Deep-Sea Research II, http://dx. ABSTRACTMulti-species statistical catch at age models (MSCAA) can quantify interacting effects of climate and fisheries harvest on species populations, and evaluate management trade-offs for fisheries that target several species in a food web. We modified an existing MSCAA model to include temperature-specific growth and predation rates and applied the modified model to three fish species, walleye pollock (Gadus chalcogrammus), Pacific cod (Gadus macrocephalus) and arrowtooth flounder (Atheresthes stomias), from the eastern Bering Sea (USA). We fit the model to data from 1979 through 2012, with and without trophic interactions and temperature effects, and use projections to derive singleand multi-species biological reference points (BRP and MBRP, respectively) for fisheries management. The multi-species model achieved a higher over-all goodness of fit to the data (i.e. lower negative log-likelihood) for pollock and Pacific cod. Variability from water temperature typically resulted in 5-15% changes in spawning, survey, and total biomasses, but did not strongly impact recruitment estimates or mortality. Despite this, inclusion of temperature in projections did have a strong effect on BRPs, including recommended yield, which were higher in single-species models for Pacific cod and arrowtooth flounder that included temperature compared to the same models without temperature effects. While the temperature-driven multi-species model resulted in higher yield MBPRs for arrowtooth flounder than the same model without temperature, we did not observe the same patterns in multi-species models for pollock and Pacific cod, where variability between harvest scenarios and predation greatly exceeded temperature-driven variability in yield MBRPs. Annual predation on juvenile pollock (primarily cannibalism) in the multi-species model was 2-5 times the annual harvest of adult fish in the system, ͵ thus predators represent a strong control on population dynamics that exceeds temperature-driven changes to growth and is attenuated through harvest-driven reductions in predator populations. Additionally, although we observed differences in spawning biomasses at the accepted biological catch (ABC) proxy between harvest scenarios and single-and multi-species models, discrepancies in spawning stock biomass estimates did not translate to large differences in yield. We found that multi-species models produced higher estimates of combined yield for aggregate maximum sustainable yield (MSY) targets than single species models, but were more conservative than singlespecies models when individual MSY targets were used, with the exception of scenarios where minimum biomass thresholds were imposed. Collectively our results suggest that climate and trophic drivers can interact to affect MBRPs, but for prey species with high predation rates, trophic-and man...

show abstract

Section: )mentioning

confidence: 99%

A comparison of fisheries biological reference points estimated from temperature-specific multi-species and single-species climate-enhanced stock assessment models

Holsman

Ianelli

Aydin

et al. 2016

Deep Sea Research Part II: Topical Studies in Oceanography

View full text Add to dashboard Cite

show abstract

“…Methods such as Stock Synthesis (Methot, 2009) and GADGET (e.g. Taylor et al, 2007) provide complex frameworks that cope with almost all data types and qualities. However, even within these comprehensive and complex methods, it is hard to control all the aspects required for a variety of multivariate data involving, agebased catch with bias, biomass from egg surveys, and total mortalities from tags, which we use here in the example of NEA mackerel.…”

Section: Introductionmentioning

confidence: 99%

Investigating agreement between different data sources using Bayesian state-space models: an application to estimating NE Atlantic mackerel catch and stock abundance

Simmonds

Portilla

Skagen³

et al. 2010

ICES Journal of Marine Science

View full text Add to dashboard Cite

Simmonds, E. J., Portilla, E., Skagen, D., Beare, D., and Reid, D. G. 2010. Investigating agreement between different data sources using Bayesian state-space models: an application to estimating NE Atlantic mackerel catch and stock abundance. – ICES Journal of Marine Science, 67: 1138–1153. Bayesian Markov chain Monte Carlo methods are ideally suited to analyses of situations where there are a variety of data sources, particularly where the uncertainties differ markedly among the data and the estimated parameters can be correlated. The example of Northeast Atlantic (NEA) mackerel is used to evaluate the agreement between available data from egg surveys, tagging, and catch-at-age using multiple models within the Bayesian framework WINBUGS. The errors in each source of information are dealt with independently, and there is extensive exploration of potential sources of uncertainty in both the data and the model. Model options include variation by age and over time of both selectivity in the fishery and natural mortality, varying the precision and calculation method for spawning-stock biomass derived from an egg survey, and the extent of missing catches varying over time. The models are compared through deviance information criterion and Bayesian posterior predictive p-values. To reconcile mortality estimated from the different datasets the landings and discards reported would have to have been between 1.7 and 3.6 times higher than the recorded catches.

show abstract

“…The flexibility of the model is obtained by permitting Gadget to estimate most of the parameters (see Stefansson (2003) and Taylor et al (2007) for more details on the iterative reweighting scheme). The inverse of these minimum sums of squared errors, multiplied by the degrees of freedom for each component, is used as a variance estimate to build the relative weight for each component.…”

Section: Weighting Procedures and Parameters Estimationmentioning

confidence: 99%

“…An appropriate integration of different components of a model (e.g. stocks, fleets), generally described by different variables from different datasets, can be done through the use of (weighted) likelihood components (Taylor et al, 2007). This also provides a basis for implementing complex and disaggregated models (Taylor and Stefansson, 2004).…”

Section: Introductionmentioning

confidence: 99%

“…Gadget is the Globally applicable Area Disaggregated General Ecosystem Toolbox developed to model marine ecosystems, and it is able to integrate species interaction and the impact of fisheries exploiting the stocks of that species (Taylor et al, 2007). As specified by Stefansson and Palsson (1998), Gadget has been developed as a forward simulation model using statistical estimation through weighted combinations of several log-likelihood criteria.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

First implementation of a Gadget model for the analysis of hake in the Mediterranean

et al. 2011

View full text Add to dashboard Cite

A simple implementation of the statistical modelling framework Gadget for cod in Icelandic waters

Cited by 23 publications

References 18 publications

A comparison of fisheries biological reference points estimated from temperature-specific multi-species and single-species climate-enhanced stock assessment models

A comparison of fisheries biological reference points estimated from temperature-specific multi-species and single-species climate-enhanced stock assessment models

Investigating agreement between different data sources using Bayesian state-space models: an application to estimating NE Atlantic mackerel catch and stock abundance

First implementation of a Gadget model for the analysis of hake in the Mediterranean

Contact Info

Product

Resources

About