Science Advisory Report  2014/060

Stock assessment and management advice for British Columbia Pacific Herring: 2014 status and 2015 forecast

Summary

• Commercial fishing for British Columbia (BC) Pacific Herring is managed as five major stock management areas: Haida Gwaii (HG), Prince Rupert District (PRD), Central Coast (CC), Strait of Georgia (SOG), and West Coast of Vancouver Island (WCVI), and two minor herring stock management areas: Haida Gwaii Area 2W, and WCVI Area 27 (Figure 1).
• The 2014 spawning biomass and forecast of the 2015 spawning biomass were assessed using an integrated statistical catch-at-age model (ISCAM or “the assessment model”).  Advice for each Pacific Herring stock is presented in probabilistic decision tables showing predicted status in 2015 given a range of constant catches relative to target harvest rates and performance metrics relating directly to the existing herring harvest control rule.
• Assessment results and advice are summarized by management area.  All herring biomass results are reported in metric tonnes (t). SB denotes spawning biomass. Stock status is measured relative to equilibrium unfished spawning biomass (SB0).  Estimates of various quantities are presented as the 5-95% credible interval, with median values in parentheses.
• There was concern that data from the Central Coast area survey resulted from an uncommonly high number of samples (6 of 13) that were collected from Statistical Area 8, which is historically an unfished area with consistently smaller (age 2) fish. As a result, Area 8 biological samples were weighted by their average relative contribution over the past 20 years (7%) in the analysis.
• On-ground observations by Nuu-chah-nulth fishers and Fisheries and Oceans Canada (DFO) staff felt that the spawning biomass in 2014 was much lower than estimated by the assessment model. In addition, it was determined that WCVI Statistical Area 24 samples were not representative and therefore removed from the analysis. 
• Two sensitivity cases were conducted: one mimicking the historical management procedure without the Poor, Average, and Good recruitment categories (fixed cut-off and assumption of q=1); and, the second a preliminary evaluation of the consequences of reductions in fishery-independent survey data on historical estimates of stock size.
• The comparison of performance of the current (base) management procedure and the former management procedure (1995-2010) illustrated the former management procedure generated lower estimates of stock biomass and lower catches at predicted harvest rates.  The historical management procedure produces lower estimates of spawning biomass for all major stocks, and higher depletion estimates (SBt /SB0) for all major stocks except for the Prince Rupert District. However, for any given catch level, the former management procedure shows there to be a higher probability of exceeding target harvest rates than for the base model.
• A ten year retrospective analysis of data for the Central Coast was conducted to examine potential implications of less frequent spawn survey data (biennial instead of annual) on historical estimates of stock size. Systematic under-estimation of spawning biomass was observed when biomass was increasing, and systematic over-estimation of spawning biomass was observed when biomass was decreasing, with catch rates exceeding the target harvest rate in the latter case. Retrospective errors can reverse abruptly as stock trends change, but the analysis could not include historical decisions that would have been made given less frequent spawn survey collections, and how the decisions would have affected the stock. However, it was agreed that the behaviour of this systematic over- and under-estimation was an undesirable property associated with less frequent surveys.

Haida Gwaii (HG)

• All herring spawning from Cumshewa Inlet in the north to Louscoone Inlet in the south are assumed to be part of the Haida Gwaii stock.
• No commercial herring fishery occurred in this area from 2005 through 2014.
• Spawn index declined from 16,025 t in 2013 to 10,566 t in 2014 and the 2014 biological samples contain a high proportion of age 4 fish and few age 3 fish.
• The estimated spawning biomass in 2014 (SB2014) is 11,912 – 43,875 t (median 22,993 t).SB2014 is estimated to be 0.35 – 1.25 (median 0.67) of SB0. The model estimates below average recruitment for 2013 and 2014.
• With low apparent recruitment entering the spawning population in 2014 and similar natural mortality estimates for recent years, the model estimates that the 2014 spawning stock biomass has declined from 2013.
• The projected spawning stock biomass in 2015, assuming no catch, is estimated to be 8,295 – 35,621 t (median 17,285 t), consisting of 5 – 31% (median 15%) age-3 fish and 46 – 86% (median 70%) age-4 and older fish.
• The median estimate of stock biomass is projected to decline in 2015 (relative to 2014), and, in the absence of fishing, the model estimates that there is a 5% probability the stock will be below the cut-off of 25% SB0 in 2015.

Prince Rupert District (PRD)

• All herring spawning in Statistical Areas 3 to 5 are assumed to belong to the Prince Rupert District stock.
• The combined total validated catch for the seine roe, the gillnet roe, and the food and bait fisheries was 2,003 t for the 2013/14 herring season. Commercial spawn-on-kelp operations also occurred in 2014.
• Spawn index declined from 25,755 t in 2013 to 17,125 t in 2014, and the 2014 biological samples contain a high proportion of age-4 fish and few age-3 fish.
• The estimated spawning biomass in 2014 (SB2014) is 15,783 – 51,772 t (median 29,023 t). SB2014 is estimated to be 0.23 - 0.84 (median 0.46) of SB0. The model estimates below average recruitment for 2013 and 2014.
• With no apparent increase in recruitment entering the spawning population in 2014 and similar natural mortality estimates for recent years, the model estimates the 2014 spawning stock biomass to be similar to the 2013 level. Recent estimates of natural mortality are highly uncertain.
• The projected spawning stock biomass in 2015, assuming no catch, is estimated to be 13,550 – 48,310 t (median 25,770 t), consisting of 6 – 20% (median 11%) age-3 fish and 66 – 89% (median 80%) age-4 and older fish. 
• The median estimate of stock biomass is projected to decline in 2015 (relative to 2014), and in the absence of fishing, there is a 12% probability the stock will be below the cut-off of 25% SB0 in 2015.

Central Coast (CC)

• All herring spawning in Kitasu Bay (a portion of Statistical Area 6), Statistical Area 7, and part of Statistical Area 8 (Kwakshua Channel and Fitzhugh Sound) are assumed to be part of the Central Coast stock.
• Commercial gillnet roe fishery caught 687 t in 2014.  No commercial herring fishery occurred in this area from 2008 through 2013. Commercial spawn-on-kelp operations also occurred in 2014.
• Spawn index declined from 20,359 t in 2013 to 13,309 t in 2014. The 2013 observation was preceded by seven years of low spawn observations (similar to the late 1960s).
• The estimated spawning biomass in 2014 (SB2014) is 14,894 – 42,791 t (median 25,384 t). SB2014 is estimated to be 0.25 - 0.69 (median 0.42) of SB0. The model estimates above average recruitment for 2010, 2012 and 2014.
• The 2014 stock assessment estimates an increase in median estimates of spawning biomass from 2012 to 2014, due in part to above average recruitment in 2010, 2012, and 2014, a high spawn index in 2013 (preceded by seven years with low index values), and apparent decreases in natural mortality.
• The projected spawning stock biomass in 2015, assuming no catch, is estimated to be 15,100 – 49,552 t (median 27,735 t), consisting of 34 – 58% (median 46%) age-3 fish and between 33 – 55% (median 44%) age-4 and older fish.
• The median estimate of stock biomass is projected to increase in 2015 (relative to 2014), and, in the absence of fishing, there is a 4% probability the stock will be below the cut-off of 25% SB0 in 2015.

Strait of Georgia (SOG)

• All herring spawning in Statistical Areas 14 to 19, 28 and 29 (excluding Section 293), and part of 13 (Herring Sections 132 and 135, Deepwater Bay area south) are assumed to belong to the Strait of Georgia herring stock.
• The combined total validated catch for the seine roe, gillnet roe, food and bait and special use fisheries was 20,307 t for the 2013/14 herring season.
• Spawn index was at near historic high levels in 2014, and the 2014 biological samples contained a high proportion of age-3 fish and similar proportions of fish ages 4-6.
• The estimated spawning biomass in 2014 (SB2014) is 116,805 – 289,976  t (median 186,273 t). SB2014 is estimated to be 0.84 – 1.96 (median 1.31) of SB0. The model estimates average recruitment in 2012-2014.
• Median spawning biomass has increased since 2010 due in part to above average recruitment in 2010 and 2011 and apparent decreases in natural mortality.
• The projected spawning stock biomass in 2015, assuming no catch, is estimated to be 108,090 – 277,900 t (median 174,350 t), consisting of 22 – 36% (median 29%) age-3 fish and 53 – 70% (median: 62%) age-4 and older fish. 
• The median estimate of stock biomass is projected to decline in 2015 (relative to 2014), and in the absence of fishing there is a 0.00% probability the stock will be below the cut-off of 25% SB0 in 2015.

West Coast Vancouver Island (WCVI)

• All herring spawning in Statistical Areas 23 to 25 are assumed to belong to the west coast of Vancouver Island herring stock.
• The WCVI stock has been closed to commercial fisheries from 2006 to 2011 and in 2013.  A commercial harvest option was available in 2012, but was not pursued.  Commercial fishing opportunities were not permitted in 2014 following an interlocutory injunction as a result of a federal court decision.
• Spawn index observations were at historically low levels between 2006 and 2008 and again in 2010 (ranging from 2,246 t to 2,875 t). Spawn index increased to 12,342 t in 2013 then to 13,901 in 2014.  Biological sampling of the WCVI stock is characterized by extremely low sample sizes in 2013 (five samples) and 2014 (three samples). The 2014 biological samples contain a high proportion of age-3 fish.
• The estimated spawning biomass in 2014 (SB2014) is 18,461 – 54,710 t (median 32,038 t). SB2014 is estimated to be 0.33 – 0.90 (median 0.55) of SB0. The model estimates below average recruitment in 2011 and 2012.
• Median spawning biomass has increased since 2008 from historically low levels, due in part to above average recruitment in 2010 and apparent decreases in natural mortality.
• The projected spawning stock biomass in 2015, assuming no catch, is estimated to be 17,549 – 56,103 t (median 31,505 t), consisting of 23 – 44% (median 33%) age-3 fish and 39 – 62% (median 52%) age-4 and older fish.
• The median estimate of stock biomass is projected to decline in 2015 (relative to 2014), and, in the absence of fishing, there is a 1% probability the stock will be below the cut-off of 25% SB0 in 2015

Area 2W

• All herring spawning in Statistical Area 2W (except Herring Section 006) are assumed to belong to this Haida Gwaii minor stock.
• A commercial spawn-on-kelp fishery occurred in this area in 2014.
• The estimated spawning biomass in 2014 (SB2014) is 1,210 – 7,847 t (median 3,268 t). SB2014 is estimated to be 0.39 – 2.17 (median 0.98) of SB0. The model estimates above average recruitment in 2013.
• Median spawning biomass has declined since 2011, due in part to decreases in the spawn index and apparent increases in natural mortality.
• The projected spawning stock biomass in 2015, assuming no catch, is estimated to be 1,083 – 8,983 t (median 3,338 t), consisting of 1 – 22% (median 6%) age-3 fish and 71– 98% (median 91%) age-4 and older fish.

Area 27

• All herring spawning in Statistical Area 27 are assumed to belong to this West Coast of Vancouver Island minor stock.
• A commercial spawn-on-kelp fishery occurred in this area in 2014.
• The estimated spawning biomass in 2014 (SB2014) is 740 – 2,509 t (median: 1,387 t). SB2014 is estimated to be 0.37 – 1.11 (median 0.66) of SB0. The model estimates below average recruitment in 2011-2013.
• Median spawning biomass has remained relatively constant over the past decade, while natural mortality may have decreased over the same period.
• The projected spawning stock biomass in 2015, assuming no catch, is 744 – 2,963 t (median 1,506 t), consisting of 12 – 61% (median 31%) age-3 fish and 31– 80% (median 58%) age-4 and older fish.  There is high uncertainty in the estimates of the number of age-3 fish in 2015.

This Science Advisory Report is from the September 3-4, 2014 Stock Assessment and Management Advice for BC Pacific Herring: 2014 Status and 2015 Forecast.  Additional publications from this meeting will be posted on the Fisheries and Oceans Canada Science Advisory Schedule as they become available.

Complete document (PDF, 628 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.