Citation
Summary
The jack mackerel stock in JMA 1 is a three-species complex made up of Trachurus novaezelandiae (JMN), T. declivis (JMD), and T. murphyi (JMM). It is targeted by a fleet of purse seine vessels that, in recent years, has primarily operated in the Bay of Plenty, although the extent of the fishery was wider before operation of the fishery was consolidated in 2018. The total allowable commercial catch (TACC) for the complex has been unchanged since 1996, when it was increased following the influx of Chilean jack mackerel (JMM). However, despite the presence of JMM being highly variable, there is currently no management strategy in place which would allow for TACC adjustments over time to reflect abundance changes in the different species present in the management complex.
Monitoring studies have been conducted to assess catch composition, including species mix, length, and age data. JMN has been the dominant species in the catch in most years since the mid-1990s. While low estimates of JMN fishing mortality from recent studies suggest less urgency for immediate catch adjustments, developing cost-effective methods to adjust catches based on population trends is essential for long-term sustainability. Currently, there is no suitable abundance index for any of the jack mackerel species in JMA 1, making it challenging to assess stock status. Composition data (length and age frequency data) have been used as performance indicators for fisheries and, especially in the case of length compositions, are relatively easy to collect.
This study investigates the evaluation of a length-based control rule as an option for adjusting JMN catch in the JMA 1 fishery. This evaluation is based on a simple metric of fishing impacts on length compositions (mean length of catch) in the context of uncertain stock dynamics and biological information. Although mean length is a basic measure, it is commonly used in stock assessment to check if models can replicate relevant composition changes and, at least in deterministic cases, can indicate changes in fishing mortality. However, the appropriateness of length-based control rules depends on various factors, including recruitment variability, fishery selectivity, and life-history assumptions. To account for uncertainties in life history and stock demography, this study fitted a flexible length- and age-based model to available data and evaluated mean-length-based harvest control rules to determine their potential impact on yield and risk when compared to a constant catch strategy.
The results indicate that the performance of these rules depends on the life-history characteristics considered. In cases with higher natural mortality, some rules provided additional yield at equivalent risk levels, but this advantage did not hold for scenarios with lower natural mortality. This dependency on life history suggests that these rules are less adequate in scenarios with high relative fishing mortality (F/Fmsy), possibly due to the high overlap of length-at-age distributions among ages selected by the fishery. This complexity in size composition makes it difficult to infer fishing mortality levels solely from mean size.
The study also highlighted uncertainties regarding the appropriateness of life-history parameters and stock structure representations in the models used. Current evidence suggests JMN exhibit fast growth and low natural mortality, characteristics that contrast with expected values for similar pelagic species. These discrepancies in parameters pose challenges for developing effective length-based control rules. Additionally, the assumption of logistic selectivity and the absence of any refuge from fishing pressure in the model led to potentially overly conservative assumptions about stock vulnerability. The lack of a mechanism for mean-size-based control rules to account for environmental variability further compromises the performance of such rules.
The findings of this evaluation contrast with previous studies that used length-based monitoring to determine catch levels. In those studies, rules based on length-based spawning potential ratio (LB-SPR) allowed for rebuilding of the population to desired levels. However, the differences between these evaluations and the present one lie in the life-history assumptions and operating models used, which strongly influence the appropriateness of management reference points and procedures. Given the unresolved nature of life-history assumptions for JMN, the study opted for a relative evaluation approach but found that sensitivity to these assumptions still impacts performance. It suggests the need for further exploration of alternative metrics and the need for a more comprehensive management strategy, based on age composition, to address the uncertainties and challenges associated with JMN management.