Age-structured Stock Assessment Research Articles

Dynamics of lake trout production in the main basin of Lake Huron

Abstract To inform lake trout (Salvelinus namaycush) fishery management in Lake Huron that has undergone rapid ecosystem changes, we quantified lake trout production dynamics by coupling age-structured stock assessment and fish bioenergetics models. Our approach revealed the connection between piscivore production and prey consumption, included growth compensation to reproduction losses, and allowed comparisons between long-term dynamics of fishery harvests and fish production. We found that, despite the collapse of alewives, a major non-native pelagic prey fish, lake trout production appeared to be sustainable. To a certain degree, the effect of recent recruitment declines on lake trout production was offset by the release of harvest pressure from subadult lake trout and reduction in fishing and sea lamprey-induced mortality on adult lake trout. Evidence for sustainability also included the finding that no changes in average ratios of annual production to beginning-of-the-year biomass. Juvenile P:B ratio remained as high as 2.1. The effect of growth declines on adult and subadult production was offset by reduction in population mortality. Body growth and condition did not continue to decline when lake trout became more and more reliant on round goby as food, and the dynamics of total consumption of prey fish continued to be recipient controlled.

Age interpretation in eulachon (Thaleichthys pacificus) as suggested by otolith microchemical signatures

Eulachon (Thaleichthys pacificus) are forage fish that play an important role in the ecosystem as prey for many species. Given their continuing population decline, determining accurate ages is necessary for age-structured stock assessments. This study employed microchemistry analysis to identify group differences and help interpret growth zone patterns on otolith surfaces. Specimens were collected off the coast of Oregon, in the coastal areas of Southeast Alaska, and in the southeastern Bering Sea. Laser-ablation inductively-coupled plasma mass spectrometry (LA-ICP-MS) was used to measure elemental ratios in the otoliths along a continuous track from the core to the proximoventral margin. Ba:Ca, Sr:Ca, Zn:Ca, and Mg:Ca signatures suggested that eulachon specimens from three geographic regions are different based on their elemental profiles. For the Oregon specimens, fluctuations in Ba:Ca and Zn:Ca signatures appeared consistent with otolith growth zones and most likely were the result of seasonal coastal upwelling events. Variations in Ba:Ca and Zn:Ca were useful as annual markers for eulachon otoliths from the Bering Sea. For the Southeast Alaska specimens, analysis of the Ba:Ca and Zn:Ca oscillations was not straightforward. Further work is needed to understand the link between otolith microchemistry, fish physiology, and regional environmental factors.

Using an integral projection model to assess the effect of temperature on the growth of gilthead seabream Sparus aurata.

Accurate information on the growth rates of fish is crucial for fisheries stock assessment and management. Empirical life history parameters (von Bertalanffy growth) are widely fitted to cross-sectional size-at-age data sampled from fish populations. This method often assumes that environmental factors affecting growth remain constant over time. The current study utilized longitudinal life history information contained in otoliths from 412 juveniles and adults of gilthead seabream, Sparus aurata, a commercially important species fished and farmed throughout the Mediterranean. Historical annual growth rates over 11 consecutive years (2002–2012) in the Gulf of Lions (NW Mediterranean) were reconstructed to investigate the effect of temperature variations on the annual growth of this fish. S. aurata growth was modelled linearly as the relationship between otolith size at year t against otolith size at the previous year t-1. The effect of temperature on growth was modelled with linear mixed effects models and a simplified linear model to be implemented in a cohort Integral Projection Model (cIPM). The cIPM was used to project S. aurata growth, year to year, under different temperature scenarios. Our results determined current increasing summer temperatures to have a negative effect on S. aurata annual growth in the Gulf of Lions. They suggest that global warming already has and will further have a significant impact on S. aurata size-at-age, with important implications for age-structured stock assessments and reference points used in fisheries.

Size-selectivity for British Columbia Sablefish (Anoplopoma fimbria) estimated from a long-term tagging study

The underlying size-distribution of commercial fish stocks is usually unknown, so fishery size-selectivity must be estimated as a latent process embedded within age-structured stock assessments. However, dome-shaped fishery size-selectivity, in particular, is often inestimable because decreasing selectivity is confounded with mortality at older ages. In this paper, we test for dome-shaped selectivity in British Columbia's Sablefish fishery using a long-term tagging data set. We incorporate alternative fishery size-selectivity assumptions within a mark-recapture framework based on an asymptotic logistic model and dome-shaped models using gamma and normal probability density functions. We also fit each model using both time-invariant and time-varying parameterizations. Our results strongly suggest dome-shaped size-selectivity for tagged-Sablefish in longline trap, longline hook, and bottom trawl fisheries. Time-varying models were generally favored over time-invariant models, although alternative time-varying models often produced similar statistical fits. Dome-shaped selectivity in longline fisheries could be a function of fishery targeting, fish movement, or by lower reporting rates for large size-classes.

Regional variation in the intensity of humpback whale predation on Pacific herring in the Gulf of Alaska

We modeled the biomass of Pacific herring (Clupea pallasii) consumed by humpback whales (Megaptera novaeangliae) to determine if whales are preventing the recovery of some herring populations in the Gulf of Alaska. We estimated consumption, by whales, of two depressed (Lynn Canal, Prince William Sound) and one robust (Sitka Sound) herring populations during fall/winter of 2007–2008 and 2008–2009. Consumption estimates relied on observations of whale abundance, prey selection, and herring energy content along with published data on whale size and metabolic rate. Herring biomass removed by whales was compared with independent estimates of herring abundance to assess the impact of predation on each population. Whales removed a greater proportion of the total biomass of herring available in Lynn Canal and Prince William Sound than in Sitka Sound. Biomass removals were greatest in Prince William Sound where we observed the largest number of whales foraging on herring. The biomass of herring consumed in Prince William Sound approximated the biomass lost to natural mortality over winter as projected by age-structured stock assessments. Though whales also focused their foraging on herring during the fall in Lynn Canal, whales were less abundant resulting in lower estimated consumption rates. Whales were more abundant in Sitka Sound than in Lynn Canal but foraged predominately on euphausiids. Herring abundance was greater in Sitka Sound, further reducing the overall impact on the herring population. These data indicate that the focused predation in Prince William Sound can exert top-down controlling pressure, but whale populations are not a ubiquitous constraint on forage fish productivity in the Gulf of Alaska at this time.

Analysis of the corvina gulf fishery as a function of management actions in the Upper Gulf of California, Mexico

The gulf corvina (Cynoscion othonopterus), is an endemic species of the Gulf of California, and its fishery is one of the most important in the Upper Gulf of California. Two dynamic biomass models (Schaefer & Pella-Tomlinson), were used to assess the state of the stock due to the lack of enough age data for a reliable full age-structured stock assessment. The models required an historical annual time-series of the abundance index (from 1993 to 2010), derived from the commercial catch and effort data. The Akaike information criterion indicates that the best model was the Schaefer model. The maximum sustainable yield (MSY), estimated with the Schaefer model was 3.100 ton, with a maximum surplus biomass (BMSY) that will allow the capture of 8.200 ton, and a fishing effort (fMSY) of 457 boats. The fishing mortality (F = 0.43) was 26% higher than the fishing mortality at the biological reference point (F0.1 = 0.34). The average biomass from 2006 to 2010 was 52% of the optimum level of the fishery (Est2006-2010 < 1). In 1999 an increase of the fishing effort accelerated the decrease of the biomass. These results indicate that the stock has not been healthy, in spite of the Biosphere Reserve decree, because the core zone has not been respected as a prohibited zone for fishing, and because of the increased fishing effort.

Interannual and Spatial Variability in Maturity of Walleye Pollock Gadus chalcogrammus and Implications for Spawning Stock Biomass Estimates in the Gulf of Alaska.

Catch quotas for walleye pollock Gadus chalcogrammus, the dominant species in the groundfish fishery off Alaska, are set by applying harvest control rules to annual estimates of spawning stock biomass (SSB) from age-structured stock assessments. Adult walleye pollock abundance and maturity status have been monitored in early spring in Shelikof Strait in the Gulf of Alaska for almost three decades. The sampling strategy for maturity status is largely characterized as targeted, albeit opportunistic, sampling of trawl tows made during hydroacoustic surveys. Trawl sampling during pre-spawning biomass surveys, which do not adequately account for spatial patterns in the distribution of immature and mature fish, can bias estimated maturity ogives from which SSB is calculated. Utilizing these maturity data, we developed mixed-effects generalized additive models to examine spatial and temporal patterns in walleye pollock maturity and the influence of these patterns on estimates of SSB. Current stock assessment practice is to estimate SSB as the product of annual estimates of numbers at age, weight at age, and mean maturity at age for 1983-present. In practice, we found this strategy to be conservative for a time period from 2003–2013 as, on average, it underestimates SSB by a 4.7 to 11.9% difference when compared to our estimates of SSB that account for spatial structure or both temporal and spatial structure. 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.

Can autocorrelated recruitment be estimated using integrated assessment models and how does it affect population forecasts?

The addition of juveniles to marine populations (termed “recruitment”) is highly variable due to variability in the survival of fish through larval and juvenile stages. Recruitment estimates are often large or small for several years in a row (termed “autocorrelated” recruitment). Autocorrelated recruitment can be due to numerous factors, but typically is attributed to multi-year environmental drivers affecting early life survival rates. Estimating the magnitude of recruitment autocorrelation within a stock assessment model and examinations on its effect on the quality of forecasts of spawning biomass within stock assessments is uncommon. We used a simulation experiment to evaluate the estimability of autocorrelation within a stock assessment model over a range of levels of autocorrelation in recruitment deviations. The precision and accuracy of estimated autocorrelation, and the ability of an integrated age-structured stock assessment framework to forecast the dynamics of the system, were compared for scenarios where the autocorrelation parameter within the assessment was fixed at zero, fixed at its true value, internally estimated within the integrated model, or input as a fixed value determined using an external estimation procedure that computed the sample autocorrelation of estimated recruitment deviations. Internal estimates of autocorrelation were biased toward extreme values (i.e., towards 1.0 when true autocorrelation was positive and −1.0 when true autocorrelation was negative). Estimates of autocorrelation obtained from the external estimation procedure were nearly unbiased. Forecast performance was poor (i.e., true biomass outside the predictive interval for the forecasted biomass) when autocorrelation was ignored, but was non-zero in the simulation. Applying the external estimation procedure generally improved forecast performance by decreasing forecast error and improving forecast interval coverage. However, estimates of autocorrelation were shown to degrade when fewer than 40 years of recruitment estimates were available.

Reply to comments by Riley and Dunlop on He et al. (2015)

He et al. (2015) described piscivory patterns in themain basin of Lake Huron for 1984–2010, during which there was also a pattern of stepwise declines in the abundance of dominant prey fish species. The approach of He et al. (2015) was to couple age-structured stock assessment and fish bioenergetics models to estimate prey fish consumption and to compare these patterns with prey fish biomass from a bottom trawl survey. Riley andDunlop (2016) were highly critical of themethods and conclusions reached byHe et al. (2015). They claimed that we incorrectly interpreted the bottom trawl survey data and did not account for uncertainty. We respond to these and other criticisms below, which we find do not undermine our findings.

The effect of length bin width on growth estimation in integrated age-structured stock assessments

Analysts conducting stock assessments using integrated, age-structured models must discretize length data into a limited number of bins (data bins). Furthermore, some modeling frameworks also allow users to specify a distinct structure for how lengths of fish are represented in the model (model bins). The effect of choices regarding the number and width of these bins on model output is unclear, and these choices are made inconsistently in assessments across regions and species. Here, we used the stock synthesis modeling framework, and the ss3sim stock assessment simulation package, to explore the effects of choices about length discretization on stock assessment performance for three fish life-history types and four data cases. We found that, with all other aspects of a model fixed, increasing the model bin width tended to increase estimates of spawning biomass, but this effect depended on the shape of length-based processes (e.g., growth, maturity, and selectivity). Thus, we suggest analysts using model bins wider than 1cm explore the effect of this decision on derived management quantities. In the context of estimation, there generally was a predictable tradeoff between estimation accuracy and model run time, with finer model and data bins always improving estimation accuracy and model convergence, but increasing run time. In some cases, wider data bins reduced run time (by up to 50%) with little sacrifice in model estimation performance, particularly those using conditional age-at-length data. This study identifies key aspects to consider when binning length, and provides pertinent information for stock assessment best practice guidelines.

Estimating fish growth for stock assessments using both age–length and tagging-increment data

In age-structured stock assessments it would be useful to be able to include all available information on growth, including age–length observations and length increments from tagging experiments. However, it was suggested in 1988 that combing the growth information from these two sources was problematic because the age- and length-based growth information they contain are not directly comparable. We evaluate two approaches that have since been made to this problem and conclude that though both approaches achieve comparability the simpler method was better suited for use in stock assessments, in part because of lesser computational demands. We show how the simpler approach is improved by allowing for correlation between length deviates at tagging and recapture, which increases biological plausibility and corrects a negative bias in estimates of variability in length at age.

Growth in age-structured stock assessment models

Growth – the way that fish get bigger as they get older – is one component of the population models used in fisheries stock assessments. I review current practice related to this component in age-structured models, discussing how growth is specified, its functions in the stock assessment model, and how growth-related data are used. I then discuss some associated problems and suggest that (a) we should be cautious about assuming that all fishery selectivities are size-based, and that an age–size sample is always random at size; (b) the jury is still out on the question of whether growth variation by phenotype is a useful model feature; (c) when growth is time-varying, conditional age-at-size data might better used outside the model (as an age–size key); and (d) that it is often difficult to extract much growth information from size composition data. I conclude by presenting a strategy for addressing the structural, procedural, and statistical decisions that must be made in finding the growth model best suited to our data.

Age estimation and lead-radium dating of golden tilefish, Lopholatilus chamaeleonticeps

There is a growing concern over the lack of life history information for many deepwater fisheries species, including golden tilefish, Lopholatilus chamaeleonticeps. Fundamental life history characteristics, like age and growth, are required for effective, age-structured stock assessments and management decisions. A previous effort to validate golden tilefish age estimates using bomb radiocarbon dating was inconclusive, which led to an application of lead-radium dating in the current study. Lead-radium dating uses the radioactive disequilibrium of lead-210 (210Pb) and radium-226 (226Ra) in otoliths as an independent estimate of age. Ages were also estimated using traditional age estimates by counting growth zones in thin otolith sections and lead-radium dating was used to test these estimates. Radiometric ages (corrected for time since capture) were similar to age estimates from growth zone counts for two of the female age groups and the two oldest age groups of unknown sex, which confirmed an annual growth zone deposition. However, radiometric ages did not agree with age estimates from growth zone counts for males. The difference may be attributed to geographical variations in radium levels, growth rates and growth zone formation by gender or gender transition. Male sagittal otoliths revealed inconsistent growth zone patterns in thin sections, which may have contributed to underageing. Golden tilefish longevity was confirmed to 26 years.

Coupling age-structured stock assessment and fish bioenergetics models: a system of time-varying models for quantifying piscivory patterns during the rapid trophic shift in the main basin of Lake Huron

We quantified piscivory patterns in the main basin of Lake Huron during 1984–2010 and found that the biomass transfer from prey fish to piscivores remained consistently high despite the rapid major trophic shift in the food webs. We coupled age-structured stock assessment models and fish bioenergetics models for lake trout (Salvelinus namaycush), Chinook salmon (Oncorhynchus tshawytscha), walleye (Sander vitreus), and lake whitefish (Coregonus clupeaformis). The model system also included time-varying parameters or variables of growth, length–mass relations, maturity schedules, energy density, and diets. These time-varying models reflected the dynamic connections that a fish cohort responded to year-to-year ecosystem changes at different ages and body sizes. We found that the ratio of annual predation by lake trout, Chinook salmon, and walleye combined with the biomass indices of age-1 and older alewives (Alosa pseudoharengus) and rainbow smelt (Osmerus mordax) increased more than tenfold during 1987–2010, and such increases in predation pressure were structured by relatively stable biomass of the three piscivores and stepwise declines in the biomass of alewives and rainbow smelt. The piscivore stability was supported by the use of alternative energy pathways and changes in relative composition of the three piscivores. In addition, lake whitefish became a new piscivore by feeding on round goby (Neogobius melanostomus). Their total fish consumption rivaled that of the other piscivores combined, although fish were still a modest proportion of their diet. Overall, the use of alternative energy pathways by piscivores allowed the increases in predation pressure on dominant diet species.

Looking in the rear-view mirror: bias and retrospective patterns in integrated, age-structured stock assessment models

Abstract Retrospective patterns are systematic changes in estimates of population size, or other assessment model-derived quantities, that occur as additional years of data are added to, or removed from, a stock assessment. These patterns are an insidious problem, and can lead to severe errors when providing management advice. Here, we use a simulation framework to show that temporal changes in selectivity, natural mortality, and growth can induce retrospective patterns in integrated, age-structured models. We explore the potential effects on retrospective patterns of catch history patterns, as well as model misspecification due to not accounting for time-varying biological parameters and selectivity. We show that non-zero values for Mohn’s ρ (a common measure for retrospective patterns) can be generated even where there is no model misspecification, but the magnitude of Mohn’s ρ tends to be lower when the model is not misspecified. The magnitude and sign of Mohn’s ρ differed among life histories, with different life histories reacting differently from each type of temporal change. The value of Mohn’s ρ is not related to either the sign or magnitude of bias in the estimate of terminal year biomass. We propose a rule of thumb for values of Mohn’s ρ which can be used to determine whether a stock assessment shows a retrospective pattern.

Assessing billfish stocks: A review of current methods and some future directions

Billfishes, excluding swordfish, are typically not the primary targets of large-scale fisheries, which has historically led to a lack of targeted monitoring efforts. The lack of data on age-composition, missing catch data and species identification problems for some species, lack of fishery-independent index data, as well as environmental influences on population dynamics is compounded by the international nature of how fisheries for billfishes are assessed and managed. This paper overviews the most recent assessments conducted for 20 stocks of billfishes in the Indian, Atlantic and Pacific Oceans, and how management advice is provided for these stocks. Assessments for billfishes are conducted using a wide range of techniques, ranging from catch-only methods which infer stock status based primarily on the prior distributions assumed for the parameters of a population dynamics model, to statistical catch-at-age analyses that integrate a wide range of data types. Key recommendations arising from this review include that age-structured stock assessments should be based on models that allow sex-structure to be represented, a full accounting for uncertainty requires adequately representing uncertainty regarding growth rates, natural mortality, the form and parameters of the stock–recruitment relationship, and how data are weighted, and that if biomass dynamics models are to be applied, they should be based on Bayesian state-space formulations rather than observation or process error estimators because such formulations are better able to represent uncertainty.

The importance of length and age composition data in statistical age-structured models for marine species

AbstractManagement of marine resources depends on the assessment of stock status in relation to established reference points. However, many factors contribute to uncertainty in stock assessment outcomes, including data type and availability, life history, and exploitation history. A simulation–estimation framework was used to examine the level of bias and accuracy in assessment model estimates related to the quality and quantity of length and age composition data across three life-history types (cod-, flatfish-, and sardine-like species) and three fishing scenarios. All models were implemented in Stock Synthesis, a statistical age-structured stock assessment framework. In general, the value of age composition data in informing estimates of virgin recruitment (R0), relative spawning-stock biomass (SSB100/SSB0), and terminal year fishing mortality rate (F100), decreased as the coefficient of variation of the relationship between length and age became greater. For this reason, length data were more informative than age data for the cod and sardine life histories in this study, whereas both sources of information were important for the flatfish life history. Historical composition data were more important for short-lived, fast-growing species such as sardine. Infrequent survey sampling covering a longer period was more informative than frequent surveys covering a shorter period.

Fishery selection and its relevance to stock assessment and fishery management

Fishery selection (selectivity for short) is the term often used to describe the phenomenon whereby a fish stock experiences mortality due to fishing that is age- or size-specific. Selectivity operates both at a local scale, as in the direct interactions of individual fish with the fishing gear (contact selection), and at a stock-wide scale (population selection), as evidenced by the differential rates of fishing mortality-at-age that are generally observed in stock assessment results. All age-structured stock assessment models have some form of fishery selection to modulate the impact of fishing mortality on differing age-classes, but selection coefficients, from a stock assessment viewpoint, generally are nuisance parameters rather than a focus of attention. This paper provides an overview of the three main processes that contribute to and influence population selection: (1) physical sorting by the fishing gear or differential behavioral responses of the fish to the gear produce the phenomenon of contact selection; (2) differing selection properties of different types of fishing gear (e.g., trawl versus longline) in turn generate a composite selection curve that is a weighted average of the different kinds of contact selection; and (3) when the fish are not well mixed spatially, then the spatial distribution of fishing relative to the spatial distribution of the fish also affects population selectivity. Fishing mortality-at-age estimates derived from a published Virtual Population Analysis of Scotian Shelf haddock are used to illustrate the diversity of shapes that can be seen in population selection curves and their considerable temporal variability. A spatial model for fishery age-selectivity is then used to demonstrate that the maximum relative yield harvested from a stock can be a function of both contact selection and the spatial distribution of fishing.

Global population trajectories of tunas and their relatives

Tunas and their relatives dominate the world's largest ecosystems and sustain some of the most valuable fisheries. The impacts of fishing on these species have been debated intensively over the past decade, giving rise to divergent views on the scale and extent of the impacts of fisheries on pelagic ecosystems. We use all available age-structured stock assessments to evaluate the adult biomass trajectories and exploitation status of 26 populations of tunas and their relatives (17 tunas, 5 mackerels, and 4 Spanish mackerels) from 1954 to 2006. Overall, populations have declined, on average, by 60% over the past half century, but the decline in the total adult biomass is lower (52%), driven by a few abundant populations. The trajectories of individual populations depend on the interaction between life histories, ecology, and fishing pressure. The steepest declines are exhibited by two distinct groups: the largest, longest lived, highest value temperate tunas and the smaller, short-lived mackerels, both with most of their populations being overexploited. The remaining populations, mostly tropical tunas, have been fished down to approximately maximum sustainable yield levels, preventing further expansion of catches in these fisheries. Fishing mortality has increased steadily to the point where around 12.5% of the tunas and their relatives are caught each year globally. Overcapacity of these fisheries is jeopardizing their long-term sustainability. To guarantee higher catches, stabilize profits, and reduce collateral impacts on marine ecosystems requires the rebuilding of overexploited populations and stricter management measures to reduce overcapacity and regulate threatening trade.

Stock assessment and projections incorporating discard estimates in some years: an application to the hake stock in ICES Divisions VIIIc and IXa

Abstract Fernández, C., Cerviño, S., Pérez, N., and Jardim, E. 2010. Stock assessment and projections incorporating discard estimates in some years: an application to the hake stock in ICES Divisions VIIIc and IXa. – ICES Journal of Marine Science, 67: 1185–1197. A Bayesian age-structured stock assessment model is developed to take into account available information on discards and to handle gaps in the time-series of discard estimates. The model incorporates mortality attributable to discarding, and appropriate assumptions about how this mortality may change over time are made. The result is a stock assessment that accounts for information on discards while, at the same time, producing a complete time-series of discard estimates. The method is applied to the hake stock in ICES Divisions VIIIc and IXa, for which the available data indicate that some 60% of the individuals caught are discarded. The stock is fished by Spain and Portugal, and for each country, there are discard estimates for recent years only. Moreover, the years for which Portuguese estimates are available are only a subset of those with Spanish estimates. Two runs of the model are performed; one assuming zero discards and another incorporating discards. When discards are incorporated, estimated recruitment and fishing mortality for young (discarded) ages increase, resulting in lower values of the biological reference points Fmax and F0.1 and, generally, more optimistic future stock trajectories under F-reduction scenarios.