Decline of West Coast Rockfishes

Rockfish Management: Are We Circling the Wagons Around the Wrong Paradigm? Rockfish species whose status has been assessed:

Black Pacific Ocean Perch Bocaccio Shortbelly Canary Shortspine Thornyhead Chilipepper Splitnose Cowcod Widow Darkblotched Yelloweye Longspine Thornyhead Yellowtail yelloweye rockfish Red = overfished (7 = 50%) Steven Berkeley, Long Marine Lab, U. C. Santa Cruz Blue = not overfished (7 = 50%) Mark Hixon, Department of Zoology, Oregon State University Total rockfish species in FMP = 63 (22% assessed) Ralph Larson, San Francisco State University Milton Love, U.C. Santa Barbara

Decline of West Coast Rockfishes Species Maximum Age Age at 50% Maturity rebuilding target year ↓ Black 40 6 2038 2030 Bocaccio 50 ? 2074 2023 Canary 84 8 2090 2027 Cowcod 55 ? 2058 yelloweye Darkblotched 105 8

Pacific Ocean 100 8 Perch Widow 60 6

Yelloweye 118 19

NOAA Fisheries

“Folly, the perverse persistence in a policy demonstrably unworkable” (Smith 1998)

Starting yearly trip limits (in pounds, and monthly equivalents) for West Coast Sebastes spp. 450,000 Current management is based on controlling 400,000 S. Mendocino fishing mortality to maintain the spawning 350,000 N. Mendocino biomass at 40% of the level that would exist 300,000 250,000 without fishing. 200,000 150,000 100,000 50,000 0 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002

1 100

80 1000

800 r

60 e 600

b m

u 400 N 40 200

0 5 B0

Percent of Virgin Stock of Virgin Percent 20 10 B 15 40 k 20 B25 oc 25 St B g A 30 10 in 0 ge 35 wn pa B0 B40 B25 B10 S Biological Reference Point

Fundamental Assumptions of Conventional Fisheries Management: For at least those stocks with highly episodic recruitment, the consequences of age-class truncation may be catastrophic, but this possibility is ignored by conventional 1. Larvae of all females have an equal probability of fish stock management techniques. survival regardless of maternal age, or time or location of spawning. Longhurst, 2002. 2. Stocks are well-mixed, so it makes no difference where fish are caught, only how many.

Old fish have longer spawning seasons:

180

160

140 14 and older 120 age 11-13

100 age 8 - 10 age 5 - 7 80

20 Relative Larval Production Larval Relative 0 3rd - Jan 4th - Jan 1st - Feb 2nd - Feb 3rd - Feb 4th - Feb 1st - Mar 2nd - Mar Week - Month

Match: larvae survive:

→ food larvae from old fish Black Rockfish

Mis-Match: larvae starve: larvae

Abundance Abundance from food young fish Bobko & Berkeley Time of year → (2004)

2 1996 30 25 age 5 - 7 age 8 - 10 20 age 11 - 13 15 14 and older 10 5 0 What about larval quality? 3rd - Jan 4th - Jan 1st - Feb 2nd - Feb 3rd - Feb 4th - Feb 1st - Mar 2nd - Mar

Percent Larval Production Larval Percent Date (Week - Month) Do older fish produce more competent larvae?

30 25 20 15 10 5 0 3rd - Jan 4th - Jan 1st - Feb 2nd - Feb 3rd - Feb 4th - Feb 1st - Mar 2nd - Mar Date (Week - Month) Percent Juvenile Birthdates

0.08

0.07

0.06

0.05

0.04

0.03 Growth = -0.15 + 0.21 (1-e-0.28 age) 0.02 p = 0.0004 r2 = 0.71 0.01 n = 16

Mean Daily Growth Rate (mm/day) Rate Growth Daily Mean 0.00 4 6 8 10 12 14 16 18 Maternal Age

12 So, all spawning biomass is not equivalent. Maternal age 10 affects the time of larval production and the length of the spawning season. 8 And all larvae are not equivalent. Larvae of older females 50% mort (days) = -8.5 + 22.5 (1-e-0.18 age) 6 p < 0.0001 are more competent and almost certainly more likely to Days Until 50% Mortality r2 = 0.73 survive. n = 21 4 4 6 8 10 12 14 16 18 Maternal Age

3 Multiple stocks per rockfish species Assumption of conventional fisheries biology: 4 population subdivisions ٭ off Oregon '01 Stocks are well-mixed. and southern Black Rockfish Washington:

identified from ٭ 02' New data: chemical signatures in daily growth rings of otoliths Stocks are actually comprised of multiple, reproductively distinct units.

otolith (ear stone) rings ٭ 02'01-'

(Miller & Shanks (2004 ٭ 02'01-'

Multiple stocks per rockfish species Multiple stocks per rockfish species

3 population 3 population subdivisions subdivisions Pacific Ocean Perch between off (overfished): Puget Sound British and Columbia: Pacific Coast, with a genetic Copper Rockfish identified from gradient along microsatellite DNA at 5 loci the open coast: identified from microsatellite DNA at 6 loci

Buonaccorsi et al. (2002) Withler et al. (2001)

But, it may be more complicated than that. Recruitment may come from restricted temporal and spatial oceanographic windows which vary from one year to the next (i.e. the sweepstakes-chance matching hypothesis, Hedgecock, 1994).

4 Shortbelly rockfish Genetic Diversity Within Samples (Nei’s H statistic: Diagonal) Ne estimates for cohort pairs Genetic Differentiation Between Samples (Nei’s G: Off-Diagonal) Cohorts Fk S Ne 95% CI Adults Larvae March Total Farallon Juveniles June June 1995-1996 0.0028 48-82 2,073 1,456-2,798 Juvs Juvs 1996-1997 -0.0002 82-123 ∞ - Adults 0.6844 1997-1998 0.0012 123-67 4,807 3,367-6,500 Larvae 0.0070 0.7355 1998-1999 -0.0015 67-627 ∞ - Mar. 0.0148 0.0161 0.6211 Juv. 1999-2000 0.0001 627-140 101,728 73,470-134,513 Tot. 0.0549* 0.0447* 0.0247 0.5887 Mean 0.0005 12,345 8,692-16,416 June Juvs. 6 1 -3 Farallon 0.0677** 0.0572** 0.0316* -- 0.5532 N= 16.7 x 10 (spawning population) ; Ne/N ≈ 10 Juvs 1. Mean estimate (1964-2003) from Rogers et al. (2000) Status of the darkblotched resource in 2000. PFMC (www.pcouncil.org) Larson, R. J, C. Orrego and R.W. Julian, 1995. (* p<0.10; ** p < 0.05) Gomez-Uchida & Banks (2004)

Fundamental Assumptions of Conventional So, what do we know that suggests a new management Fisheries Management: approach may be needed?

For at least some species, we know: 1. Larvae of all females have an equal probability of 1. Older fish spawn earlier than younger fish. survival regardless of maternal age, or time or location 2. Larvae of older females are more likely to survive. of spawning. 3. Stocks are much more complex than previously thought so local depletions might not be replenished from other areas. 4. Recruitment may come from distinct time/area windows that 2. Stocks are well-mixed, so it makes no difference where vary from year to year. fish are caught, only how many.

Looking only at how different fishing strategies So, what do we do about these findings? Can affect age structure which in turn affects larval conventional management approaches survival: accommodate this new information and if not, 1. B what tools are most appropriate? 40 2. Slot limit w/25% release mortality 3. 20% marine reserve 4. Reduced fishing mortality rate

5 Mgmt Spawning % Virgin % Yield % of Strategy Age Classes SSB Relative to Virgin Age

B40 12+ SSB

B0 29 100 0 100 Management Strategy Effective Larval Output B 40 15 40 100 12.5 (F=0.177) (% of B40) Slot (w/25% 19 40 94 18.3 B0 314 rel. mort)

20% MR (F B40 100 outside = 29 40 96 22.3 0.316) Slot (w/25% release mort.) 103 F adj to MR on table 17 50 85 22.3 20% MR on table 107 (F=0.124) F adj to 20% MR on table 132 20% MR off table (F 29 52 80 30.0 20% MR off table 143 outside = 0.177) F adj to MR SSB off table 154 F adj to MR off table 19 56 75 30.0 (F=0.0982)

Conclusions:

Mgmt Strategy Recruitment Yield Fishing Marine Slot Limit (% of B ) (% of B ) 40 40 Mortality Reserves

B40 (F=0.177) 100 100 Time of Maybe Yes Probably not Slot w/25% release spawning 103 97 mort. Age Maybe Yes Probably not 20% MR on table 107 103 structure

F adj to 20% MR on 107 103 Stock spatial No Yes (if No table (F=0.1635) complexity networked) 20% MR off table 143 115 Hedgecock No Yes (if No F adj to 20% MR off 143 115 effect networked) table (F=0.110)