The estimated impact of changes to otolith field-sampling and ageing effort on stock assessment inputs, outputs, and catch advice

Abstract

Generating accurate data for stock assessments is resource-demanding, necessitating periodic evaluation of survey sampling designs and potential impacts on stock assessments. We developed a framework for bootstrapped resampling of survey age data and calculation of input sample sizes as a function of among-bootstrap variance in age compositions. Data from this bootstrap estimator were then used to evaluate the influence of alternative sampling rates and methods on uncertainty in estimates of the overfishing limit (OFL) calculated using stock assessment models. For dusky rockfish and Pacific ocean perch, a 10% decrease in the number of tows sampled upon led to a predicted 5-6% increase in the CV of OFL (log-log slope = -0.576 to -0.486), which was greater than the 0-2% increase from a 10% decrease in otoliths-per-tow (log-log slope = -0.238 to -0.029). Application of this approach across all stocks monitored in the survey of interest is required to identify which stocks (i) benefit the most from increased sampling of ageing structures, or (ii) cost the least in terms of OFL uncertainty owing to reduced sampling.

Reductions in sampling effort for fishery-independent age and length composition: balancing sampling efficiency, data uncertainty, and workforce health

“…To determine whether and when age composition ISS reaches an asymptote, simulation studies might be more appropriate rather than bootstrapping historical data, as variables such as life history or intrahaul correlation can be accounted for specifically (e.g., Hulson et al 2011;Xu et al 2020). The results of this study and Siskey et al (2023) suggest that the number of age samples necessary to maximize age composition information are likely beyond current management agency capacities. Our recommendation, therefore, is to attempt to retain, at a minimum, historical levels of sampling for age composition, as reductions in these sampling rates have downstream effects on assessment model uncertainty.…”

“…As we increased the subsampling rates for age composition, the age composition ISS continued to increase, indicating that a sample size producing diminishing returns had not been reached, unlike the asymptote observed for length composition ISS. In a similar study, Siskey et al (2023) evaluated the effect of increased age sampling beyond current sampling levels for a subset of stocks and found that the age composition ISS continued to increase, up to age sampling levels 67% greater than current rates. Siskey et al (2023) also found that, while holding the total age samples constant, improvements to age composition ISS were achieved by increasing sampling in more hauls even while decreasing the number of specimen age data collected in any given haul.…”

“…The number of hauls in which these stocks were present, and the number of those hauls sampled for length frequency and age specimen data, are shown in Table S1. It was noted in Siskey et al (2023) that, for some of these stocks, there were some age-length pairs in the AFSC age specimen data where the length was not reflected in the length frequency data. This also occurred in the Can.…”

“…It is recognized in sampling theory that increasing the number of samples increases the precision of the estimate, and from this perspective, one would expect that the maximum number of samples per each survey observation would increase precision. Improvements in precision of age or length composition estimates are not only influenced by the magnitude of sampling, but also the way samples are collected, for example, considering the number of samples collected within a haul versus changing the number of hauls from which samples were obtained (Siskey et al 2023). Furthermore, each fish sampled on a survey has an immediate physical labor cost (e.g., at-sea time management issues), a cumulative health cost (e.g., repetitive stress injuries), downstream labor costs (e.g., staff effort reading otoliths to determine age), and budget costs (e.g., the monetary cost to perform a fisheries-independent survey).…”

“…In fisheries stock assessments, this realized sample size is used within the multinomial likelihood to fit age and (or) length composition data, termed "input sample size (ISS)" to distinguish that it is being used as an input to a stock assessment model (Thorson et al 2023). Several studies have used ISS to evaluate impacts on assessment results, including whether ISS can be estimated within the assessment model (Hulson et al 2012;Francis 2017;Thorson et al 2017), and more recently as a tool to evaluate the effect of changes to sampling methodologies on assessment model uncertainty (Siskey et al 2023). Furthermore, bootstrapping methods that mimic the sampling designs for length and age composition have been developed to estimate ISS external to an assessment model (Stewart and Hamel 2014).…”