James Thorson - Academia.edu (original) (raw)

Papers by James Thorson

Information regarding several intensively managed groundfish off the U.S. West Coast is obtained ... more Information regarding several intensively managed groundfish off the U.S. West Coast is obtained from a randomized bottom trawl survey. However, bottom trawl tows are sometimes fouled by bottom structures that may be associated with higher densities of these target species, and trawl performance for non-fouled tows may also be affected by bottom structures. Indices of abundance resulting from this bottom trawl may be suspect due to these trawl performance issues, which has prompted the development of visual sampling methods that can access untrawlable habitats. In this study, we use a spatial simulation model representing habitat selection for individuals and shoals of a Pacific rockfish (Sebastes spp.) to evaluate different sampling designs that could combine data from visual and bottom trawl sampling gears into a single index of abundance. Specifically, we explore simple and stratified sampling designs involving both gears. We explore stratification based on an estimated index of ...

ICES Journal of Marine Science

Methods for determining appropriate management actions for data-poor stocks, including annual cat... more Methods for determining appropriate management actions for data-poor stocks, including annual catch limits (ACLs), have seen an explosion of research interest in the past decade. We perform an inventory of methods for determining ACLs for stocks in the United States, and find that ACLs are assigned to 371 stocks and/or stock complexes with 193 (52%) determined using methods involving catch data only. The proportion of ACLs involving these methods varies widely among fisheries management regions, with all the 67 ACLs in the Caribbean determined using recent catch when compared with 1 of 33 ACLs in the New England region (US Northeast). Given this prevalence of data-poor ACLs, we recommend additional research regarding the potential effectiveness of simple management procedures for data-poor stocks that are currently managed using ACLs. In particular, simple management procedures may allow a broader range of data types and management instruments that better suit the particulars of ind...

Stock assessments typically involve a workflow where spatially referenced data are pre-processed ... more Stock assessments typically involve a workflow where spatially referenced data are pre-processed to estimate spatially aggregated measures of a population (i.e. indices of abundance, composition summaries), and these aggregate measures are then used to estimate parameters for a non-spatial population model. However, improvements in statistical and computational methods allow population dynamcis to be estimated using spatiotemporal models, which replace variables representing total abundance with random fields representing population densities over the population’s range. We provide a brief introduction to spatiotemporal methods, including their estimation using Template Model Builder, while demonstrating a new state-space spatiotemporal model involving Gompertz-form density dependence. We use simulation experiments and data for three rockfishes in the California Current to contrast spatial and non-spatial state-space models, and results indicate that non-spatial models can result in...

Fisheries Research, 2014

Integrated assessment models frequently track population abundance at age, and hence account for ... more Integrated assessment models frequently track population abundance at age, and hence account for fishery removals using a function representing fishery selectivity at age. However, fishery selectivity may have an unusual shape that does not match any parametric function. For this reason, previous research has developed flexible 'non-parametric' models for selectivity that specify a penalty on changes in selectivity as a function of age. In this study, we describe an alternative 'semi-parametric' approach to selectivity, which specifies a penalty on differences between estimated selectivity at age and a prespecified parametric model whose parameters are freely estimated, while also using cross-validation to select the magnitude of penalty in both semi-and non-parametric models. We then compare parametric, semi-parametric, and non-parametric models using simulated data and evaluate the bias and precision of estimated depletion and fishing intensity. Results show that semi-and non-parametric models result in little decrease in precision relative to the parametric model when the parametric model matches the true data-generating process, but that the semi-and non-parametric models have less bias and greater precision when the parametric function is misspecified. As expected, the semi-parametric model reverts to its pre-specified parametric form when age-composition sample size is low but performs similarly to the non-parametric model when sample size is high. Overall, results indicate few disadvantages to using the non-parametric model given the range of simulation scenarios explored here, and that the semi-parametric model provides a selectivity specification that is intermediate between parametric and non-parametric forms.

Background/Question/Methods Understanding the density dependence of fish reproduction is critical... more Background/Question/Methods Understanding the density dependence of fish reproduction is critically important for management and conservation of commercial fisheries. Stock-recruitment relationships describe how reproductive output changes relative to the number of fish in a population. Common models used to fit fisheries data include the Beverton-Holt model, which describes a system where total recruitment levels off at higher spawner densities (called compensation), and the Ricker model, where total recruitment declines at high densities (overcompensation). While large amounts of data allow an appropriate model to be chosen for a species or population, no clear method has been developed to objectively determine the appropriate stock-recruitment relationship for commercial species where limited or no stock-recruitment data exist. To address this gap, we developed a hierarchical model that uses Bayesian inference to link stock-recruitment parameters among species, taxonomic orders (...

The application of stock assessments to fisheries management on the west coast of the United Stat... more The application of stock assessments to fisheries management on the west coast of the United States has been greatly enhanced by the ongoing development of Stock Synthesis, a flexible statistical catch-at-age framework. While this framework has been enhanced to incorporate more data and parameters, recent develops have also demonstrated its use in data-limited situations. Both catch-only and catch-index only modelling exercises have proven useful in obtaining management and/or biological quantities. This work attempts to expand the catch-index applications by incorporating current year length or age compositions so as to define selectivity curves and improve biomass and status estimation. This work is applied to several west coast groundfish stocks and compared to both the full assessments and the simpler applications in order to identify the use of such data. The results aim to improve our understanding of modelling under data-constraining situations, as well as offer insight in da...

Fisheries Research, 2015

ABSTRACT Fisheries scientists are increasingly concerned about changes in vital rates caused by e... more ABSTRACT Fisheries scientists are increasingly concerned about changes in vital rates caused by environmental change and fishing impacts. Demographic parameters representing individual growth, maturity, mortality, and recruitment have previously been documented to change over decadal time scales. However, there has been relatively little comparison regarding which vital rates cause relatively greater or lesser impacts on commonly used fisheries management targets. We therefore use a life table (based on age-structured assessment models) to explore the sensitivity of fishing mortality, spawning biomass, and catch targets to changes in parameters representing growth, mortality, recruitment, and maturation rates for three representative life histories representing long-, medium-, and short-lived species. The elasticity analysis indicates that demographic changes can result in substantial variation in fisheries management targets, but that changes in mortality rates are particularly important for spawning biomass and catch targets while maturity and recruitment compensation are also important for fishing mortality targets. We conclude by discussing the importance of improved data repositories to address covariation among maturity, growth, and mortality parameters.

Fisheries Research, 2014

Please cite this article in press as: Thorson, J.T., Cope, J.M., Catch curve stock-reduction anal... more Please cite this article in press as: Thorson, J.T., Cope, J.M., Catch curve stock-reduction analysis: An alternative solution to the catch equations. Fish. Res. (2014), http://dx.

Ecological Applications, 2015

Identifying spatiotemporal hotspots is important for understanding basic ecological processes, bu... more Identifying spatiotemporal hotspots is important for understanding basic ecological processes, but is particularly important for species at risk. A number of terrestrial and aquatic species are indirectly affected by anthropogenic impacts, simply because they tend to be associated with species that are targeted for removals. Using newly developed statistical models that allow for the inclusion of time-varying spatial effects, we examine how the co-occurrence of a targeted and nontargeted species can be modeled as a function of environmental covariates (temperature, depth) and interannual variability. The nontarget species in our case study (eulachon) is listed under the U.S. Endangered Species Act, and is encountered by fisheries off the U.S. West Coast that target pink shrimp. Results from our spatiotemporal model indicated that eulachon bycatch risk decreases with depth and has a convex relationship with sea surface temperature. Additionally, we found that over the 2007-2012 period, there was support for an increase in eulachon density from both a fishery data set (+40%) and a fishery-independent data set (+55%). Eulachon bycatch has increased in recent years, but the agreement between these two data sets implies that increases in bycatch are not due to an increase in incidental targeting of eulachon by fishing vessels, but because of an increasing population size of eulachon. Based on our results, the application of spatiotemporal models to species that are of conservation concern appears promising in identifying the spatial distribution of environmental and anthropogenic risks to the population.

Methods in Ecology and Evolution, 2015

ABSTRACT Predicting and explaining the distribution and density of species is one of the oldest c... more ABSTRACT Predicting and explaining the distribution and density of species is one of the oldest concerns in ecology. Species distributions can be estimated using geostatistical methods, which estimate a latent spatial variable explaining observed variation in densities, but geostatistical methods may be imprecise for species with low densities or few observations. Additionally, simple geostatistical methods fail to account for correlations in distribution among species, and generally estimate such cross-correlations as a post-hoc exercise.We therefore present spatial factor analysis (SFA), a spatial model for estimating a low-rank approximation to multivariate data, and use it to jointly estimate the joint distribution of multiple species simultaneously. We also derive an analytic estimate of cross-correlations among species from SFA parameters.As a first example, we show that distributions for 10 bird species in the breeding bird survey in 2013 can be parsimoniously represented using only 5 spatial factors. As a second case study, we show that forward-prediction of catches for 20 rockfishes (Sebastes spp.) off the U.S. West Coast is more accurate using spatial factor analysis than analyzing each species individually. Finally, we show that single-species models give a different picture of cross-correlations than joint estimation using SFA.SFA complements a growing list of tools for jointly modelling the distribution of multiple species, and provides a parsimonious summary of cross-correlation without requiring explicit declaration of habitat variables. We conclude by proposing future research that would model species cross-correlations using dissimilarity of species’ traits, and the development of spatial dynamic factor analysis for a low-rank approximation to spatial time-series data.This article is protected by copyright. All rights reserved.

ABSTRACT Meta-analyses of stock assessments can provide novel insight into the population dynamic... more ABSTRACT Meta-analyses of stock assessments can provide novel insight into the population dynamics and ecology of fished species. Controlled manipulation of fish populations in open marine systems is rarely possible, but fisheries data provide a valuable substitute for such manipulations. The RAM Legacy Stock Assessment Database facilitates meta-analysis of commercially exploited marine fishes and invertebrates by providing time series of total biomass, spawner biomass, recruitment, fishing mortality and catch/landings for more than 350 stocks in a common format. Recent analyses of the database have shown that: (1) productivity of fish stocks is frequently better explained by a series of “productivity regimes” than by simple production models; (2) there is little or no evidence for depensation across the range of stock sizes observed for fish populations in the database; (3) the failure of many fish stocks to recover can generally be explained by insufficient reductions in fishing mortality. Taken together, these results offer support for the standard (non-depensatory) population models used to predict stock recovery, but also provide a caution about the large uncertainty introduced by a changing marine environment.

Marine fish populations have high variation in cohort strength, and the production of juveniles (... more Marine fish populations have high variation in cohort strength, and the production of juveniles (recruitment) may have persistent positive or negative residuals (autocorrelation) after accounting for spawning biomass. Autocorrelated recruitment will occur whenever average recruitment levels change between oceanographic regimes or due to predator release, but may also indicate persistent environmental and biological effects on shorter time-scales. Here, we use estimates of recruitment variability and autocorrelation to simulate the stationary distribution of spawning biomass for 100 real-world stocks when unfished, fished at F MSY , or fished following a harvest control rule where fishing mortality decreases as a function of spawning biomass. Results show that unfished stocks have spawning biomass (SB) below its deterministic equilibrium value (SB 0 ) 58% of the time, and below 0.5SB 0 5% of the time on average across all stocks. Similarly, stocks fished at the level producing deterministic maximum sustainable yield (F MSY ) are below its deterministic prediction of spawning biomass (SB MSY ) 60% of the time and below 0.5SB MSY 8% of the time. These probabilities are greater for stocks with high recruitment variability, positive autocorrelation, and high natural mortality-traits that are particularly associated with clupeids and scombrids. An elevated probability of stochastic depletion, i.e. biomass below the deterministic equilibrium expectation, implies that management actions required when biomass drops below a threshold may be triggered more frequently than expected. Therefore, we conclude by suggesting that fisheries scientists routinely calculate these probabilities during stock assessments as a decision support tool for fisheries managers.

Canadian Journal of Fisheries and Aquatic Sciences, 2014

Accurate estimates of abundance are imperative for successful conservation and management. Classi... more Accurate estimates of abundance are imperative for successful conservation and management. Classical, stratified abundance estimators provide unbiased estimates of abundance, but such estimators may be imprecise and impede assessment of population status and trend when the distribution of individuals is highly variable in space. Model-based procedures that account for important environmental covariates can improve overall precision, but frequently there is uncertainty about the contribution of particular environmental variables and a lack of information about variables that are important determinants of abundance. We develop a general semiparametric mixture model that incorporates measured habitat variables and a nonparametric smoothing term to account for unmeasured variables. We contrast this spatial habitat approach with two stratified abundance estimators and compare the three models using an intensively managed marine fish, darkblotched rockfish (Sebastes crameri). We show that the spatial habitat model yields more precise, biologically reasonable, and interpretable estimates of abundance than the classical methods. Our results suggest that while design-based estimators are unbiased, they may exaggerate temporal variability of populations and strongly influence inference about population trend. Furthermore, when such estimates are used in broader meta-analyses, such imprecision may affect the broader biological inference (e.g., the causes and consequences of the variability of populations).

Ecology, 2014

Count data arise frequently in ecological analyses, but regularly violate the equi-dispersion con... more Count data arise frequently in ecological analyses, but regularly violate the equi-dispersion constraint imposed by the most popular distribution for analyzing these data, the Poisson distribution. Several approaches for addressing over-dispersion have been developed (e.g., negative binomial distribution), but methods for including both underdispersion and over-dispersion have been largely overlooked. We provide three specific examples drawn from life-history theory, spatial ecology, and community ecology, and illustrate the use of the Conway-Maxwell-Poisson (CMP) distribution as compared to other common models for count data. We find that where equi-dispersion is violated, the CMP distribution performs significantly better than the Poisson distribution, as assessed by information criteria that account for the CMP's additional distribution parameter. The Conway-Maxwell-Poisson distribution has seen rapid development in other fields such as risk analysis and linguistics, but is relatively unknown in the ecological literature. In addition to providing a more flexible exponential distribution for count data that is easily integrated into generalized linear models, the CMP allows ecologists to focus on the magnitude of under-or over-dispersion as opposed to the simple rejection of the equi-dispersion null hypothesis. By demonstrating its suitability in a variety of common ecological applications, we hope to encourage its wider adoption as a flexible alternative to the Poisson.

Oikos, 2014

ABSTRACT Short-term forecasts based on time series of counts or survey data are widely used in po... more ABSTRACT Short-term forecasts based on time series of counts or survey data are widely used in population biology to provide advice concerning the management, harvest and conservation of natural populations. A common approach to produce these forecasts uses time-series models, of different types, fit to time series of counts. Similar time-series models are used in many other disciplines, however relative to the data available in these other disciplines, population data are often unusually short and noisy and models that perform well for data from other disciplines may not be appropriate for population data. In order to study the performance of time-series forecasting models for natural animal population data, we assembled 2379 time series of vertebrate population indices from actual surveys. Our data were comprised of three vastly different types: highly variable (marine fish productivity), strongly cyclic (adult salmon counts), and small variance but long-memory (bird and mammal counts). We tested the predictive performance of 49 different forecasting models grouped into three broad classes: autoregressive time-series models, non-linear regression-type models and non-parametric time-series models. Low-dimensional parametric autoregressive models gave the most accurate forecasts across a wide range of taxa; the most accurate model was one that simply treated the most recent observation as the forecast. More complex parametric and non-parametric models performed worse, except when applied to highly cyclic species. Across taxa, certain life history characteristics were correlated with lower forecast error; specifically, we found that better forecasts were correlated with attributes of slow growing species: large maximum age and size for fishes and high trophic level for birds.

Ecology, 2014

Identifying the existence and magnitude of density dependence is one of the oldest concerns in ec... more Identifying the existence and magnitude of density dependence is one of the oldest concerns in ecology. Ecologists have aimed to estimate density dependence in population and community data by fitting a simple autoregressive (Gompertz) model for density dependence to time series of abundance for an entire population. However, it is increasingly recognized that spatial heterogeneity in population densities has implications for population and community dynamics. We therefore adapt the Gompertz model to approximate, local densities over continuous space instead of population-wide abundance, and allow productivity to vary spatially using Gaussian random fields. We then show that the conventional (nonspatial) Gompertz model can result in biased estimates of density dependence (e.g., identifying oscillatory dynamics when not present) if densities vary spatially. By contrast, the spatial Gompertz model provides accurate and precise estimates of density dependence for a variety of simulation scenarios and data availabilities. These results are corroborated when comparing spatial and nonspatial models for data from 10 years and -100 sampling stations for three long-lived rockfishes (Sebastes spp.) off the California, USA coast. In this case, the nonspatial model estimates implausible oscillatory dynamics on an annual time scale, while the spatial model estimates strong autocorrelation and is supported by model selection tools. We conclude by discussing the importance of improved data archiving techniques, so that spatial models can be used to reexamine classic questions regarding the existence and magnitude of density. dependence in wild populations.

Ecology, 2014

ABSTRACT Managing natural populations and communities requires detailed information regarding dem... more ABSTRACT Managing natural populations and communities requires detailed information regarding demographic processes at large spatial and temporal scales. This combination is challenging for both traditional scientific surveys, which often operate at localized scales, and recent citizen science designs, which often provide data with few auxiliary information (i.e., no information about individual age or condition). We therefore combine citizen science data at large scales with the demographic resolution afforded by recently developed, site-structured demographic models. We apply this approach to categorical data generated from citizen science representing species density of two managed reef fishes in the Gulf of Mexico, and use a modified Dail-Madsen model to estimate demographic trends, habitat associations, and interannual variability in recruitment. This approach identifies strong preferences for artificial structure for the recovering Goliath grouper, while revealing little evidence of either habitat associations or trends in abundance for mutton snapper. Results are also contrasted with a typical generalized linear mixed-model (GLMM) approach, using real-world and simulated data, to demonstrate the importance of accounting for the statistical complexities implied by spatially structured citizen science data. We conclude by discussing the increasing potential for synthesizing demographic models and citizen science data, and the management benefits that can be accrued.

Transactions of the American Fisheries Society, 2011

Catch-curve analysis is one of the simplest methods for stock assessment and is widely applied in... more Catch-curve analysis is one of the simplest methods for stock assessment and is widely applied in data-poor fisheries. Conventional catch-curve methods rely on the strong assumptions of constant fishing and natural mortality rates above some fully selected age that is usually estimated by visually inspecting a plot of catch at age. Here, we evaluate the performance of three catch-curve methods that relax or modify these assumptions by (1) estimating logistic selectivity parameters, (2) assuming Lorenzen-form natural mortality (natural mortality that decreases with weight), and (3) using both methods simultaneously. We used simulation modeling and decision tables to compare estimates of fishing mortality from four catch-curve methods, including the conventional method, across a variety of observable and unobservable data characteristics. We then applied the methods to catch-at-age data for Atlantic menhaden Brevoortia tyrannus from the U.S. South Atlantic fishery management region and compared the resulting estimates with published estimates of fishing mortality (F). In our simulation modeling, catch curves that estimated logistic selectivity parameters performed better than those derived by the conventional method when logistic selectivity was present. There was generally little difference in performance between estimates assuming constant natural mortality and those assuming Lorenzen natural mortality. The improvements from estimating selectivity parameters were particularly pronounced when the sample sizes for catch-at-age data were large: in those instances, estimating selectivity improved the estimation accuracy for F by nearly 20%. In our example involving Atlantic menhaden, estimates of F assuming logistic selectivity were most similar to those of published stock assessments, which had previously estimated logistic selectivity at age. We recommend our logistic-selectivity catch curve when selectivity is likely to be logistic because it improves accuracy at only a very small cost in terms of computational complexity.

Reviews in Fisheries Science, 2009

Catchability is an important parameter in many stock assessment models because it relates an inde... more Catchability is an important parameter in many stock assessment models because it relates an index of abundance to stock size. We review the theory and evidence for time-varying catchability, its effects on stock assessment estimates, and methods to include time-varying catchability in stock assessments. Numerous studies provide strong evidence that timevarying catchability is common in most fisheries and many fishery-independent surveys and can be caused by anthropogenic, environmental, biological, and management processes. Trends in catchability over time can cause biased estimates of stock size and fishing mortality rates in stock assessment models that do not compensate for them. Methods that use descriptive and functional relationships have been developed to incorporate time-varying catchability in stock assessment models. We recommend that the default assumption for stock assessments should be that catchability varies over time and that multiple methods of including time-varying catchability should be applied. Additional studies are needed to determine relative performance of alternative methods and to develop methods for selecting among models.

PLoS ONE, 2012

Concerns over fishing impacts on marine populations and ecosystems have intensified the need to i... more Concerns over fishing impacts on marine populations and ecosystems have intensified the need to improve ocean management. One increasingly popular market-based instrument for ecological stewardship is the use of certification and eco-labeling programs to highlight sustainable fisheries with low environmental impacts. The Marine Stewardship Council (MSC) is the most prominent of these programs. Despite widespread discussions about the rigor of the MSC standards, no comprehensive analysis of the performance of MSC-certified fish stocks has yet been conducted. We compared status and abundance trends of 45 certified stocks with those of 179 uncertified stocks, finding that 74% of certified fisheries were above biomass levels that would produce maximum sustainable yield, compared with only 44% of uncertified fisheries. On average, the biomass of certified stocks increased by 46% over the past 10 years, whereas uncertified fisheries increased by just 9%. As part of the MSC process, fisheries initially go through a confidential pre-assessment process. When certified fisheries are compared with those that decline to pursue full certification after pre-assessment, certified stocks had much lower mean exploitation rates (67% of the rate producing maximum sustainable yield vs. 92% for those declining to pursue certification), allowing for more sustainable harvesting and in many cases biomass rebuilding. From a consumer's point of view this means that MSC-certified seafood is 3-5 times less likely to be subject to harmful fishing than uncertified seafood. Thus, MSC-certification accurately identifies healthy fish stocks and conveys reliable information on stock status to seafood consumers.

Information regarding several intensively managed groundfish off the U.S. West Coast is obtained ... more Information regarding several intensively managed groundfish off the U.S. West Coast is obtained from a randomized bottom trawl survey. However, bottom trawl tows are sometimes fouled by bottom structures that may be associated with higher densities of these target species, and trawl performance for non-fouled tows may also be affected by bottom structures. Indices of abundance resulting from this bottom trawl may be suspect due to these trawl performance issues, which has prompted the development of visual sampling methods that can access untrawlable habitats. In this study, we use a spatial simulation model representing habitat selection for individuals and shoals of a Pacific rockfish (Sebastes spp.) to evaluate different sampling designs that could combine data from visual and bottom trawl sampling gears into a single index of abundance. Specifically, we explore simple and stratified sampling designs involving both gears. We explore stratification based on an estimated index of ...

ICES Journal of Marine Science

Methods for determining appropriate management actions for data-poor stocks, including annual cat... more Methods for determining appropriate management actions for data-poor stocks, including annual catch limits (ACLs), have seen an explosion of research interest in the past decade. We perform an inventory of methods for determining ACLs for stocks in the United States, and find that ACLs are assigned to 371 stocks and/or stock complexes with 193 (52%) determined using methods involving catch data only. The proportion of ACLs involving these methods varies widely among fisheries management regions, with all the 67 ACLs in the Caribbean determined using recent catch when compared with 1 of 33 ACLs in the New England region (US Northeast). Given this prevalence of data-poor ACLs, we recommend additional research regarding the potential effectiveness of simple management procedures for data-poor stocks that are currently managed using ACLs. In particular, simple management procedures may allow a broader range of data types and management instruments that better suit the particulars of ind...

Stock assessments typically involve a workflow where spatially referenced data are pre-processed ... more Stock assessments typically involve a workflow where spatially referenced data are pre-processed to estimate spatially aggregated measures of a population (i.e. indices of abundance, composition summaries), and these aggregate measures are then used to estimate parameters for a non-spatial population model. However, improvements in statistical and computational methods allow population dynamcis to be estimated using spatiotemporal models, which replace variables representing total abundance with random fields representing population densities over the population’s range. We provide a brief introduction to spatiotemporal methods, including their estimation using Template Model Builder, while demonstrating a new state-space spatiotemporal model involving Gompertz-form density dependence. We use simulation experiments and data for three rockfishes in the California Current to contrast spatial and non-spatial state-space models, and results indicate that non-spatial models can result in...

Fisheries Research, 2014

Integrated assessment models frequently track population abundance at age, and hence account for ... more Integrated assessment models frequently track population abundance at age, and hence account for fishery removals using a function representing fishery selectivity at age. However, fishery selectivity may have an unusual shape that does not match any parametric function. For this reason, previous research has developed flexible 'non-parametric' models for selectivity that specify a penalty on changes in selectivity as a function of age. In this study, we describe an alternative 'semi-parametric' approach to selectivity, which specifies a penalty on differences between estimated selectivity at age and a prespecified parametric model whose parameters are freely estimated, while also using cross-validation to select the magnitude of penalty in both semi-and non-parametric models. We then compare parametric, semi-parametric, and non-parametric models using simulated data and evaluate the bias and precision of estimated depletion and fishing intensity. Results show that semi-and non-parametric models result in little decrease in precision relative to the parametric model when the parametric model matches the true data-generating process, but that the semi-and non-parametric models have less bias and greater precision when the parametric function is misspecified. As expected, the semi-parametric model reverts to its pre-specified parametric form when age-composition sample size is low but performs similarly to the non-parametric model when sample size is high. Overall, results indicate few disadvantages to using the non-parametric model given the range of simulation scenarios explored here, and that the semi-parametric model provides a selectivity specification that is intermediate between parametric and non-parametric forms.

Background/Question/Methods Understanding the density dependence of fish reproduction is critical... more Background/Question/Methods Understanding the density dependence of fish reproduction is critically important for management and conservation of commercial fisheries. Stock-recruitment relationships describe how reproductive output changes relative to the number of fish in a population. Common models used to fit fisheries data include the Beverton-Holt model, which describes a system where total recruitment levels off at higher spawner densities (called compensation), and the Ricker model, where total recruitment declines at high densities (overcompensation). While large amounts of data allow an appropriate model to be chosen for a species or population, no clear method has been developed to objectively determine the appropriate stock-recruitment relationship for commercial species where limited or no stock-recruitment data exist. To address this gap, we developed a hierarchical model that uses Bayesian inference to link stock-recruitment parameters among species, taxonomic orders (...

The application of stock assessments to fisheries management on the west coast of the United Stat... more The application of stock assessments to fisheries management on the west coast of the United States has been greatly enhanced by the ongoing development of Stock Synthesis, a flexible statistical catch-at-age framework. While this framework has been enhanced to incorporate more data and parameters, recent develops have also demonstrated its use in data-limited situations. Both catch-only and catch-index only modelling exercises have proven useful in obtaining management and/or biological quantities. This work attempts to expand the catch-index applications by incorporating current year length or age compositions so as to define selectivity curves and improve biomass and status estimation. This work is applied to several west coast groundfish stocks and compared to both the full assessments and the simpler applications in order to identify the use of such data. The results aim to improve our understanding of modelling under data-constraining situations, as well as offer insight in da...

Fisheries Research, 2015

ABSTRACT Fisheries scientists are increasingly concerned about changes in vital rates caused by e... more ABSTRACT Fisheries scientists are increasingly concerned about changes in vital rates caused by environmental change and fishing impacts. Demographic parameters representing individual growth, maturity, mortality, and recruitment have previously been documented to change over decadal time scales. However, there has been relatively little comparison regarding which vital rates cause relatively greater or lesser impacts on commonly used fisheries management targets. We therefore use a life table (based on age-structured assessment models) to explore the sensitivity of fishing mortality, spawning biomass, and catch targets to changes in parameters representing growth, mortality, recruitment, and maturation rates for three representative life histories representing long-, medium-, and short-lived species. The elasticity analysis indicates that demographic changes can result in substantial variation in fisheries management targets, but that changes in mortality rates are particularly important for spawning biomass and catch targets while maturity and recruitment compensation are also important for fishing mortality targets. We conclude by discussing the importance of improved data repositories to address covariation among maturity, growth, and mortality parameters.

Fisheries Research, 2014

Please cite this article in press as: Thorson, J.T., Cope, J.M., Catch curve stock-reduction anal... more Please cite this article in press as: Thorson, J.T., Cope, J.M., Catch curve stock-reduction analysis: An alternative solution to the catch equations. Fish. Res. (2014), http://dx.

Ecological Applications, 2015

Identifying spatiotemporal hotspots is important for understanding basic ecological processes, bu... more Identifying spatiotemporal hotspots is important for understanding basic ecological processes, but is particularly important for species at risk. A number of terrestrial and aquatic species are indirectly affected by anthropogenic impacts, simply because they tend to be associated with species that are targeted for removals. Using newly developed statistical models that allow for the inclusion of time-varying spatial effects, we examine how the co-occurrence of a targeted and nontargeted species can be modeled as a function of environmental covariates (temperature, depth) and interannual variability. The nontarget species in our case study (eulachon) is listed under the U.S. Endangered Species Act, and is encountered by fisheries off the U.S. West Coast that target pink shrimp. Results from our spatiotemporal model indicated that eulachon bycatch risk decreases with depth and has a convex relationship with sea surface temperature. Additionally, we found that over the 2007-2012 period, there was support for an increase in eulachon density from both a fishery data set (+40%) and a fishery-independent data set (+55%). Eulachon bycatch has increased in recent years, but the agreement between these two data sets implies that increases in bycatch are not due to an increase in incidental targeting of eulachon by fishing vessels, but because of an increasing population size of eulachon. Based on our results, the application of spatiotemporal models to species that are of conservation concern appears promising in identifying the spatial distribution of environmental and anthropogenic risks to the population.

Methods in Ecology and Evolution, 2015

ABSTRACT Predicting and explaining the distribution and density of species is one of the oldest c... more ABSTRACT Predicting and explaining the distribution and density of species is one of the oldest concerns in ecology. Species distributions can be estimated using geostatistical methods, which estimate a latent spatial variable explaining observed variation in densities, but geostatistical methods may be imprecise for species with low densities or few observations. Additionally, simple geostatistical methods fail to account for correlations in distribution among species, and generally estimate such cross-correlations as a post-hoc exercise.We therefore present spatial factor analysis (SFA), a spatial model for estimating a low-rank approximation to multivariate data, and use it to jointly estimate the joint distribution of multiple species simultaneously. We also derive an analytic estimate of cross-correlations among species from SFA parameters.As a first example, we show that distributions for 10 bird species in the breeding bird survey in 2013 can be parsimoniously represented using only 5 spatial factors. As a second case study, we show that forward-prediction of catches for 20 rockfishes (Sebastes spp.) off the U.S. West Coast is more accurate using spatial factor analysis than analyzing each species individually. Finally, we show that single-species models give a different picture of cross-correlations than joint estimation using SFA.SFA complements a growing list of tools for jointly modelling the distribution of multiple species, and provides a parsimonious summary of cross-correlation without requiring explicit declaration of habitat variables. We conclude by proposing future research that would model species cross-correlations using dissimilarity of species’ traits, and the development of spatial dynamic factor analysis for a low-rank approximation to spatial time-series data.This article is protected by copyright. All rights reserved.

ABSTRACT Meta-analyses of stock assessments can provide novel insight into the population dynamic... more ABSTRACT Meta-analyses of stock assessments can provide novel insight into the population dynamics and ecology of fished species. Controlled manipulation of fish populations in open marine systems is rarely possible, but fisheries data provide a valuable substitute for such manipulations. The RAM Legacy Stock Assessment Database facilitates meta-analysis of commercially exploited marine fishes and invertebrates by providing time series of total biomass, spawner biomass, recruitment, fishing mortality and catch/landings for more than 350 stocks in a common format. Recent analyses of the database have shown that: (1) productivity of fish stocks is frequently better explained by a series of “productivity regimes” than by simple production models; (2) there is little or no evidence for depensation across the range of stock sizes observed for fish populations in the database; (3) the failure of many fish stocks to recover can generally be explained by insufficient reductions in fishing mortality. Taken together, these results offer support for the standard (non-depensatory) population models used to predict stock recovery, but also provide a caution about the large uncertainty introduced by a changing marine environment.

Marine fish populations have high variation in cohort strength, and the production of juveniles (... more Marine fish populations have high variation in cohort strength, and the production of juveniles (recruitment) may have persistent positive or negative residuals (autocorrelation) after accounting for spawning biomass. Autocorrelated recruitment will occur whenever average recruitment levels change between oceanographic regimes or due to predator release, but may also indicate persistent environmental and biological effects on shorter time-scales. Here, we use estimates of recruitment variability and autocorrelation to simulate the stationary distribution of spawning biomass for 100 real-world stocks when unfished, fished at F MSY , or fished following a harvest control rule where fishing mortality decreases as a function of spawning biomass. Results show that unfished stocks have spawning biomass (SB) below its deterministic equilibrium value (SB 0 ) 58% of the time, and below 0.5SB 0 5% of the time on average across all stocks. Similarly, stocks fished at the level producing deterministic maximum sustainable yield (F MSY ) are below its deterministic prediction of spawning biomass (SB MSY ) 60% of the time and below 0.5SB MSY 8% of the time. These probabilities are greater for stocks with high recruitment variability, positive autocorrelation, and high natural mortality-traits that are particularly associated with clupeids and scombrids. An elevated probability of stochastic depletion, i.e. biomass below the deterministic equilibrium expectation, implies that management actions required when biomass drops below a threshold may be triggered more frequently than expected. Therefore, we conclude by suggesting that fisheries scientists routinely calculate these probabilities during stock assessments as a decision support tool for fisheries managers.

Canadian Journal of Fisheries and Aquatic Sciences, 2014

Accurate estimates of abundance are imperative for successful conservation and management. Classi... more Accurate estimates of abundance are imperative for successful conservation and management. Classical, stratified abundance estimators provide unbiased estimates of abundance, but such estimators may be imprecise and impede assessment of population status and trend when the distribution of individuals is highly variable in space. Model-based procedures that account for important environmental covariates can improve overall precision, but frequently there is uncertainty about the contribution of particular environmental variables and a lack of information about variables that are important determinants of abundance. We develop a general semiparametric mixture model that incorporates measured habitat variables and a nonparametric smoothing term to account for unmeasured variables. We contrast this spatial habitat approach with two stratified abundance estimators and compare the three models using an intensively managed marine fish, darkblotched rockfish (Sebastes crameri). We show that the spatial habitat model yields more precise, biologically reasonable, and interpretable estimates of abundance than the classical methods. Our results suggest that while design-based estimators are unbiased, they may exaggerate temporal variability of populations and strongly influence inference about population trend. Furthermore, when such estimates are used in broader meta-analyses, such imprecision may affect the broader biological inference (e.g., the causes and consequences of the variability of populations).

Ecology, 2014

Count data arise frequently in ecological analyses, but regularly violate the equi-dispersion con... more Count data arise frequently in ecological analyses, but regularly violate the equi-dispersion constraint imposed by the most popular distribution for analyzing these data, the Poisson distribution. Several approaches for addressing over-dispersion have been developed (e.g., negative binomial distribution), but methods for including both underdispersion and over-dispersion have been largely overlooked. We provide three specific examples drawn from life-history theory, spatial ecology, and community ecology, and illustrate the use of the Conway-Maxwell-Poisson (CMP) distribution as compared to other common models for count data. We find that where equi-dispersion is violated, the CMP distribution performs significantly better than the Poisson distribution, as assessed by information criteria that account for the CMP's additional distribution parameter. The Conway-Maxwell-Poisson distribution has seen rapid development in other fields such as risk analysis and linguistics, but is relatively unknown in the ecological literature. In addition to providing a more flexible exponential distribution for count data that is easily integrated into generalized linear models, the CMP allows ecologists to focus on the magnitude of under-or over-dispersion as opposed to the simple rejection of the equi-dispersion null hypothesis. By demonstrating its suitability in a variety of common ecological applications, we hope to encourage its wider adoption as a flexible alternative to the Poisson.

Oikos, 2014

ABSTRACT Short-term forecasts based on time series of counts or survey data are widely used in po... more ABSTRACT Short-term forecasts based on time series of counts or survey data are widely used in population biology to provide advice concerning the management, harvest and conservation of natural populations. A common approach to produce these forecasts uses time-series models, of different types, fit to time series of counts. Similar time-series models are used in many other disciplines, however relative to the data available in these other disciplines, population data are often unusually short and noisy and models that perform well for data from other disciplines may not be appropriate for population data. In order to study the performance of time-series forecasting models for natural animal population data, we assembled 2379 time series of vertebrate population indices from actual surveys. Our data were comprised of three vastly different types: highly variable (marine fish productivity), strongly cyclic (adult salmon counts), and small variance but long-memory (bird and mammal counts). We tested the predictive performance of 49 different forecasting models grouped into three broad classes: autoregressive time-series models, non-linear regression-type models and non-parametric time-series models. Low-dimensional parametric autoregressive models gave the most accurate forecasts across a wide range of taxa; the most accurate model was one that simply treated the most recent observation as the forecast. More complex parametric and non-parametric models performed worse, except when applied to highly cyclic species. Across taxa, certain life history characteristics were correlated with lower forecast error; specifically, we found that better forecasts were correlated with attributes of slow growing species: large maximum age and size for fishes and high trophic level for birds.

Ecology, 2014

Identifying the existence and magnitude of density dependence is one of the oldest concerns in ec... more Identifying the existence and magnitude of density dependence is one of the oldest concerns in ecology. Ecologists have aimed to estimate density dependence in population and community data by fitting a simple autoregressive (Gompertz) model for density dependence to time series of abundance for an entire population. However, it is increasingly recognized that spatial heterogeneity in population densities has implications for population and community dynamics. We therefore adapt the Gompertz model to approximate, local densities over continuous space instead of population-wide abundance, and allow productivity to vary spatially using Gaussian random fields. We then show that the conventional (nonspatial) Gompertz model can result in biased estimates of density dependence (e.g., identifying oscillatory dynamics when not present) if densities vary spatially. By contrast, the spatial Gompertz model provides accurate and precise estimates of density dependence for a variety of simulation scenarios and data availabilities. These results are corroborated when comparing spatial and nonspatial models for data from 10 years and -100 sampling stations for three long-lived rockfishes (Sebastes spp.) off the California, USA coast. In this case, the nonspatial model estimates implausible oscillatory dynamics on an annual time scale, while the spatial model estimates strong autocorrelation and is supported by model selection tools. We conclude by discussing the importance of improved data archiving techniques, so that spatial models can be used to reexamine classic questions regarding the existence and magnitude of density. dependence in wild populations.

Ecology, 2014

ABSTRACT Managing natural populations and communities requires detailed information regarding dem... more ABSTRACT Managing natural populations and communities requires detailed information regarding demographic processes at large spatial and temporal scales. This combination is challenging for both traditional scientific surveys, which often operate at localized scales, and recent citizen science designs, which often provide data with few auxiliary information (i.e., no information about individual age or condition). We therefore combine citizen science data at large scales with the demographic resolution afforded by recently developed, site-structured demographic models. We apply this approach to categorical data generated from citizen science representing species density of two managed reef fishes in the Gulf of Mexico, and use a modified Dail-Madsen model to estimate demographic trends, habitat associations, and interannual variability in recruitment. This approach identifies strong preferences for artificial structure for the recovering Goliath grouper, while revealing little evidence of either habitat associations or trends in abundance for mutton snapper. Results are also contrasted with a typical generalized linear mixed-model (GLMM) approach, using real-world and simulated data, to demonstrate the importance of accounting for the statistical complexities implied by spatially structured citizen science data. We conclude by discussing the increasing potential for synthesizing demographic models and citizen science data, and the management benefits that can be accrued.

Transactions of the American Fisheries Society, 2011

Catch-curve analysis is one of the simplest methods for stock assessment and is widely applied in... more Catch-curve analysis is one of the simplest methods for stock assessment and is widely applied in data-poor fisheries. Conventional catch-curve methods rely on the strong assumptions of constant fishing and natural mortality rates above some fully selected age that is usually estimated by visually inspecting a plot of catch at age. Here, we evaluate the performance of three catch-curve methods that relax or modify these assumptions by (1) estimating logistic selectivity parameters, (2) assuming Lorenzen-form natural mortality (natural mortality that decreases with weight), and (3) using both methods simultaneously. We used simulation modeling and decision tables to compare estimates of fishing mortality from four catch-curve methods, including the conventional method, across a variety of observable and unobservable data characteristics. We then applied the methods to catch-at-age data for Atlantic menhaden Brevoortia tyrannus from the U.S. South Atlantic fishery management region and compared the resulting estimates with published estimates of fishing mortality (F). In our simulation modeling, catch curves that estimated logistic selectivity parameters performed better than those derived by the conventional method when logistic selectivity was present. There was generally little difference in performance between estimates assuming constant natural mortality and those assuming Lorenzen natural mortality. The improvements from estimating selectivity parameters were particularly pronounced when the sample sizes for catch-at-age data were large: in those instances, estimating selectivity improved the estimation accuracy for F by nearly 20%. In our example involving Atlantic menhaden, estimates of F assuming logistic selectivity were most similar to those of published stock assessments, which had previously estimated logistic selectivity at age. We recommend our logistic-selectivity catch curve when selectivity is likely to be logistic because it improves accuracy at only a very small cost in terms of computational complexity.

Reviews in Fisheries Science, 2009

Catchability is an important parameter in many stock assessment models because it relates an inde... more Catchability is an important parameter in many stock assessment models because it relates an index of abundance to stock size. We review the theory and evidence for time-varying catchability, its effects on stock assessment estimates, and methods to include time-varying catchability in stock assessments. Numerous studies provide strong evidence that timevarying catchability is common in most fisheries and many fishery-independent surveys and can be caused by anthropogenic, environmental, biological, and management processes. Trends in catchability over time can cause biased estimates of stock size and fishing mortality rates in stock assessment models that do not compensate for them. Methods that use descriptive and functional relationships have been developed to incorporate time-varying catchability in stock assessment models. We recommend that the default assumption for stock assessments should be that catchability varies over time and that multiple methods of including time-varying catchability should be applied. Additional studies are needed to determine relative performance of alternative methods and to develop methods for selecting among models.

PLoS ONE, 2012

Concerns over fishing impacts on marine populations and ecosystems have intensified the need to i... more Concerns over fishing impacts on marine populations and ecosystems have intensified the need to improve ocean management. One increasingly popular market-based instrument for ecological stewardship is the use of certification and eco-labeling programs to highlight sustainable fisheries with low environmental impacts. The Marine Stewardship Council (MSC) is the most prominent of these programs. Despite widespread discussions about the rigor of the MSC standards, no comprehensive analysis of the performance of MSC-certified fish stocks has yet been conducted. We compared status and abundance trends of 45 certified stocks with those of 179 uncertified stocks, finding that 74% of certified fisheries were above biomass levels that would produce maximum sustainable yield, compared with only 44% of uncertified fisheries. On average, the biomass of certified stocks increased by 46% over the past 10 years, whereas uncertified fisheries increased by just 9%. As part of the MSC process, fisheries initially go through a confidential pre-assessment process. When certified fisheries are compared with those that decline to pursue full certification after pre-assessment, certified stocks had much lower mean exploitation rates (67% of the rate producing maximum sustainable yield vs. 92% for those declining to pursue certification), allowing for more sustainable harvesting and in many cases biomass rebuilding. From a consumer's point of view this means that MSC-certified seafood is 3-5 times less likely to be subject to harmful fishing than uncertified seafood. Thus, MSC-certification accurately identifies healthy fish stocks and conveys reliable information on stock status to seafood consumers.