Science Response 2016/052

Stock Assessment and Management Advice for BC Pacific Herring: 2016 Status and 2017 Forecast

Context

Fisheries and Oceans Canada (DFO) Pacific Fisheries Management Branch requested that DFO Pacific Science Branch assess the status of British Columbia (BC) herring stocks in 2016, and provide projections of potential herring abundance in 2017 and the consequences of a range of potential harvests to inform the development of the 2016/17 Integrated Fisheries Management Plan (IFMP).

Pacific Herring abundance is currently assessed using a statistical catch-age model. The catch-age model is fitted to commercial catch data, proportions-at-age data and a fishery-independent spawning biomass index to estimate biomass and recruitment and to generate 1-year forecasts of spawning biomass (Martell et al. 2012; DFO 2015a). Seven versions of the model are fitted, respectively, to data for five major stocks: Haida Gwaii (HG), Prince Rupert District (PRD), Central Coast (CC), Strait of Georgia (SOG) and West Coast Vancouver Island (WCVI), and two minor stocks: Area 2W and Area 27. A revised catch-age model was introduced for BC herring assessments in 2006 (Haist and Schweigert 2006), and the design of the model has since undergone several iterations that have re-structured various model components and addressed issues identified during peer-review.  One major change introduced in 2011 (Martell et al. 2012) was setting the model to estimate the spawn dive survey scaling parameter q2, rather than setting it fixed at q2=1.0, as was done in previous assessment models. Another major change introduced in 2011 was to make the fishery cut-offs in the harvest control rule dependent on the model’s most recent estimate of unfished spawning biomass SB0 (i.e., cease fishing when the stock is estimated to be below 0.25SB0). In previous model iterations, the fishery cut-offs were fixed at absolute biomass levels estimated in 1996 (Schweigert et al. 1997).  Throughout this document, the term Assessment Model 1 (AM1) describes the more recent management procedure (MP), which estimates the scaling factor for the surface survey q1 (1951-1987) and dive survey q2 (1988-2016) using informative priors; and uses estimated fishery cut-offs. Assessment Model 2 (AM2) refers to an approximation of the historical MP, in which the surface survey q1 (1951-1987) is estimated, the dive survey q2 (1988-2016) is fixed at 1.0 and the fishery cut-offs are fixed at 1996 levels.

There have been a number of requests to evaluate the potential consequences of applying AM1 vs. AM2 using simulation modelling. This reflects concerns that the consequences of applying AM1 were not simulation-tested prior to its implementation in 2011, which, along with lack of rebuilding in some areas, has led to questioning the performance of AM1. In May 2015, a closed loop simulation tool was developed to evaluate performance of herring MPs against a suite of conservation and fishery performance metrics. The simulation tool was reviewed in a May 2015 Canadian Science Advisory Secretariat (CSAS) Regional Peer Review Process (DFO 2015b) and accepted as a “proof of concept” that the simulation methodology was scientifically defensible. This simulation tool was used in the development of the 2015 Science Response (DFO 2015a) in order to identify tradeoffs between MPs that assume alternative ecological hypotheses about future conditions (e.g., future patterns of natural mortality and growth for herring) and assessment modeling assumptions (e.g., the proportion of the spawn observed by the survey). Approximations of AM1 and AM2, as well as several other candidate MPs, were evaluated against a range of conservation and fishery performance criteria, including mean catch, annual average variability in catch, and the probability of dropping below candidate limit reference points 0.25SB0, 0.30SB0 (Pikitch et al. 2012) and 0.40SB0 (Pikitch et al. 2012). Results, however, were not conclusive in identifying any one procedure as generally preferable based on all metrics and trade-offs. See DFO 2015a Part 2 for a summary of the simulation results.

Continuing disagreement on the application of AM1 over AM2, and differences of opinion among First Nations, DFO resource managers, DFO Science, and industry, motivated the formation of a Pilot Technical Working Group in 2015, which was formalized as the Herring Technical Working Group (HTWG) in 2016. The HTWG consists of DFO Science and Fisheries Management and technical representatives nominated by several coastal First Nations and the herring industry, providing technical support for the development of this Science Response. A full description of members’ involvement appears at the end of this document. Results from both AM1 and AM2 were included in the 2015 Science Response (DFO 2015a), and the HTWG recommended both be included in the present document.

Fisheries and Oceans Canada (DFO) Pacific Fisheries Management Branch has requested that DFO Pacific Science Branch assess the status of BC herring stocks in 2016, and provide projections of potential herring abundance in 2017 and the consequences of a range of potential harvests to support the development of the 2016/17 Integrated Fisheries Management Plan (IFMP). The status of BC herring stocks in 2016 and forecasts for 2017 are provided in the form of dual stock assessment updates, using the AM1 (Martell et al. 2012) and AM2 (approximation of Cleary and Schweigert 2011) MPs. Both MPs have been peer reviewed through CSAS and both have been implemented in the provision of science advice for Pacific Herring in previous years. To address concerns arising both from previous CSAS processes and from implementation of each approach, the HTWG has also recommended inclusion of a table to describe the main attributes and limitations of AM1 and AM2, to support short-term decision-making (Table A.1).

Current stock status and trends, as well as projected biomass for 2017 are presented. Biomass estimates and decision tables show results from both the AM1 and AM2 MPs.

The objectives of this Science Response are to:

1. Assess the current status of Pacific Herring for each of the five major and two minor stocks using AM1 (estimate q1 for surface survey period;estimate q2 for dive survey period; use estimated cut-offs) and AM2 (estimate q1 for surface survey period; fix q2=1.0 for dive survey period; use fixed 1996 cut-offs).
2. Present trends in herring biomass, depletion, and recruitment for each major and minor stock using both AM1 and AM2 MPs.
3. Present probabilities of spawning biomass levels below cut-offs and probabilities of harvest rates exceeding targets prescribed by both AM1 and AM2, for a range of 2017 total allowable catch (TAC) levels.

Additional reference points and performance metrics are also included for the CC. These arose from discussions within the Heiltsuk-DFO Technical Team in 2015.

This Science Response Report results from the Science Response Process of September 2016 on Stock Assessment and Management Advice for BC Pacific Herring: 2016 Status and 2017 Forecast.

Complete document (PDF, 2,292 KB)

Accessibility Notice

This document is available in PDF format. If the 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.