Science Advisory Report  2015/062

Candidate Limit Reference Points as a basis for choosing among alternative Harvest Control Rules for Pacific Herring (Clupea pallasii) in British Columbia


  • British Columbia's Pacific Herring fisheries are managed based on a harvest strategy initially designed in 1986. Since adopting the strategy, two major herring stocks have remained above the management level, permitting commercial fisheries (known as the cut-off), while three stocks have been below the cut-off level in 32%, 21%, and 46% of years between 1986 and 2013.
  • Significant decreases in body size have been observed for all BC herring stocks from the early-1980s to 2010; as well as an increase in estimated natural mortality rates in some stocks since 1951 (DFO 2015). The consequences of these time-varying trends in fish stock productivity have not previously been evaluated for Pacific Herring fishery management.
  • A review of the literature did not reveal any universally accepted forage fish candidate limit reference points (LRPs) that would apply to BC herring; thus, a suite of candidates were evaluated.
  • A closed-loop simulation approach was used to explore candidate LRPs as a basis for evaluating the performance of several management procedures (the combination of data, assessment model, and harvest control rule). Simulations considered alternative theories of herring stock productivity and dynamics, as well as survey frequency (annual or biennial). This represents the first step in a management strategy evaluation (MSE) process that develops the analytical framework for future evaluations and explores the performance of some initial candidate LRPs for the five major Pacific Herring stocks.
  • The key components of the analytical framework for Pacific Herring are:
    • operating models for each of the five major Pacific Herring stocks, that represent a range of hypotheses about future changes in growth, natural mortality, and non-stationary productivity dynamics;
    • management procedures providing options in frequency of data collection, stock assessment methods/assumptions, and harvest control rules; and
    • performance indicators for comparing simulated outcomes against a suite of LRPs.
  • Management procedure performance was evaluated with respect to potential trends in growth and natural mortality using an age-structured operating model that represented four alternative scenarios of future dynamics of Pacific Herring natural mortality and growth. Natural mortality for each stock area was either held constant at 2013 values or allowed to increase by 50% over the projection period; and, growth was held constant at average 2013 values or trended toward the average historical growth rate for the period 1951-1955. 
  • Four harvest control rules were evaluated in this analysis. Two harvest control rules were developed to approximate the current practice for Fisheries and Oceans Canada (DFO) management (the difference between the two being biennial or annual spawn surveys) and two harvest control rules were based on proposed conservation rules for forage fish species.
  • Three commonly accepted fishery performance metrics were selected to provide an indication of the conservation and yield performance of simulated management procedures.  Conservation performance was measured using the probability (Pcons) of the spawning stock biomass being below the conservation threshold defined by each of the candidate LRPs.
  • A limited number of stock productivity scenarios, management procedures, and candidate LRPs were explored for the five major Pacific Herring stock areas. Nonetheless, results derived from the analytical framework provide proof of the MSE concept for evaluating LRPs, and provide a tool that can be used to update the existing management framework for BC Pacific Herring in accordance with Canada’s Sustainable Fisheries Framework.
  • The analytical framework has flexibility to explore a broader set of management procedures, ecological hypotheses and First Nations, industry, and other stakeholder objectives.
  • The objective of this analysis was not to identify a single LRP for herring management, but to identify characteristics of LRPs that appear to be most informative for assessing the ability of alternative management procedures to achieve management objectives; based on an evaluation of the stock status relative to the candidate LRPs.  LRPs based on FMSY, non-stationary unfished biomass, and the lowest biomass from which the stock has recovered, were evaluated to be less informative than fixed (equilibrium) LRPs for this purpose.  
  • The tested approximation of the current DFO management procedure appears to maintain stocks above the candidate LRPs over a narrow range of assumptions. Increases in natural mortality, similar to those estimated over the past several decades, revealed potential conservation and fishery risks in four of five stocks areas. Specifically:
    • Simulated spawning stock biomass (SSB) for Strait of Georgia (SOG) herring was maintained above the 0.25B0 LRP 95% of the time under the DFO1 and DFO2 management procedures, which is consistent with the original evaluation performed for the stock in 1988.
    • The simulations for Prince Rupert District (PRD), the Central Coast (CC) and Haida Gwaii (HG) stocks suggest that both the DFO and the Lenfest management procedures could result in SSB frequently dropping below the LRPs used in this simulation exercise
    • The LRP that was allowed to change with natural mortality and growth rates (NSB0) often failed to indicate risk in situations where risks could actually be significant. This was also true for the empirical LRP that reflected the ‘worst-case’ scenario (HistoricalB), which was lowest historical biomass from which the stock has recovered.
  • To update the Pacific Herring management framework, it is recommended that more realistic representations of the management procedures be developed for evaluation, using the simulation approach and analytical framework reviewed here. For example, consideration should be given to modelling of sequential fisheries based on multiple fleets, uncertainty in the spawn survey scaling parameter (q), and changes in size at age.

This Science Advisory Report is from the May 27 – 28, 2015 Candidate Limit Reference Points for Pacific Herring in British Columbia using a Closed-loop Simulation Modelling Approach.  Additional publications from this meeting will be posted on the Fisheries and Oceans Canada (DFO) Science Advisory Schedule as they become available.

 Complete PDF document
(18 pages; 629 K)

Accessibility Notice:

This document is available in PDF format. If the following document is not accessible to you, please contact the Secretariat to obtain another appropriate format, such as regular print, large print, Braille or audio version.

Date modified: