Wetzel et al.: The effect of reduced data on monitoring overfished fish stocks 
191 
Table 1 
Life-history and observation parameters used in the operating model and their treatment within the estima¬ 
tion method to simulate a rockfish life-history type common to the west coast of the United States. 
Parameter 
Time-invariant 
Time-varying 
Treatment in 
estimation method 
Natural mortality ( M) per year 
0.08 
0.08 
Fixed 
Natural mortality standard error (o m ) 
0 
0.10 
Natural mortality autocorrelation (p) 
0 
0.707 
Steepness (h) 
0.65 
0.65 
Estimated 
Maximum length (L„)(cm) 
64 
64 
Estimated 
Growth coefficient ( K) 
0.05 
0.05 
Estimated 
Weight at length tcj = aLP (kg) 
a=0.50xl0 -5 , 
a=1.50xl0 -5 , 
(3=3 
P=3 
Fixed 
Length at 50% maturity (cm) 
37 
37 
Fixed 
Recruitment variation (Or) 
0.50 
0.50 
Fixed 
Fishery CPUE standard error (o f ) 
0.30 
0.30 
Fixed 
Fishery CPUE catchability coefficient (Q f) 
0.01 
0.01 
Analytically estimated 
Width at maximum selectivity (cm) 
-3 
-3 
Estimated 
Width at maximum selectivity standard error (o w ) 
0 
0.20 
Size at maximum selectivity (cm) 
45 
45 
Estimated 
Size at maximum selectivity standard error (o s ) 
0 
0.05 
rockfish (Sebastes pinniger ; Thorson and Wetzel * * * 4 ), and 
Pacific ocean perch (Sebastes alutus\ Hamel and Ono 5 ). 
The reduction of fishery catch, and of the resulting 
fishery data during rebuilding, presents a challenge 
for assessment and management of rebuilding stocks. 
Many species of rockfish off the U.S. west coast (e.g., 
cowcod, yelloweye rockfish) are not reliably sampled by 
the main fishery-independent survey, the NOAA North¬ 
west Fisheries Science Center’s West Coast groundfish 
bottom trawl survey, either because of the inability of 
the survey to sample rocky habitat with trawl gear or 
because of other restrictions on sampling locations. 
Because these species are not well sampled, the ma¬ 
jority of historical information (e.g., length and age 
data) available for assessment comes primarily from 
recreational and commercial fishery samples. Yet, be¬ 
cause of restrictions on retention of fish triggered by 
rebuilding plans, often, recreational and commercial 
fishery behavior has been profoundly altered (Stewart 
et al. 2 ). In the most recent assessment of yelloweye 
rockfish, limited fishery data during rebuilding were 
cited as a challenge to “produce conclusive information 
tus and productivity of Cowcod, Sebastes levis, in the South¬ 
ern California Bight, 2013, 166 p. Fish Ecol. Div., South¬ 
west Fish. Sci. Cent., Natl. Mar. Fish. Serv., NOAA, Santa 
Cruz, CA. [Available from website.] 
4 Thorson, J. T., and C. Wetzel. 2016. The status of canary 
rockfish (Sebastes pinniger) in the California Current in 
2015, 241 p. Northwest Fish. Sci. Cent., Natl. Mar. Fish. 
Serv., NOAA, Seattle, WA. [Available from website.] 
5 Hamel, O. S., and K. Ono. 2011. Stock assessment of Pa¬ 
cific ocean perch in waters off of the U.S. West Coast in 2011, 
135 p. Pacific Fishery Management Council, Portland, 
OR. [Available from website.] 
about the stock for the foreseeable future” (Stewart et 
al. 2 ). Another overfished rockfish species, cowcod, was 
assessed most recently by using a data-moderate ap¬ 
proach that did not include length or age data instead 
of the historical data-rich integrated assessment be¬ 
cause of lack of data during the rebuilding period. Of¬ 
ten fishermen avoid targeting stocks during rebuilding 
efforts for rockfish species even as harvest limits in¬ 
crease with rebuilding populations and this precaution 
results in harvests that are well below the rebuilding 
harvest limits and in continued low levels of biological 
samples from the fishery catch. Additionally, harvest 
restrictions can affect the harvest of more abundant 
fish stocks that co-occur with rebuilding stocks and can 
result in reduced data availability that extends beyond 
a single species. 
Despite a limited harvest of a stock, continued 
data collection is necessary to determine the extent to 
which that stock has rebuilt. The ability to measure 
the rate of recovery is crucial for management, and 
increased uncertainty due to limited data can impede 
the determination of whether a stock is on track to re¬ 
build in a specified time frame. Additionally, biological 
data are critical for improvement of estimates of key 
parameters within stock assessments (e.g., natural 
mortality; growth; recruitment compensation, which 
is termed steepness) and can indicate incoming poor 
or strong recruitment year classes that will affect es¬ 
timates of relative stock biomass (the ratio of current 
biomass to unfished biomass) and rebuilding rates. 
Potential improvements in parameter estimates and 
the ability to detect incoming fluctuations in recruit¬ 
ment during rebuilding are restricted when collection 
