Catch Sampling
and Estimation in
the Federal Groundfish
Fisheries Off Alaska
1
Stratification of Federal groundfish and halibut IFQ fishing activity
as defined by regulation and the Annual Deployment Plan Process
2) Between trip
sample
2 variance
Randomization of observer deployments
within each sampled stratum:
1st level of hierarchical sample
• The collection of high quality data on both landed and at-sea discards of
federally managed species is critical for implementing the Magnuson-Stevens
Fishery Conservation and Act standards, particularly National Standard 1
and associated guidelines that require the total accounting of catch.
• Data sources are consolidated for catch estimation in a
database called the Alaska Region Catch Accounting System
(CAS); sampling and computational details are presented to
the right.
4
Catch estimates by
haul, species, retained & discard
for observed trips
Complete shoreside
enumeration of King and
Chum salmon from
observed trips,
proportionally distributed
to individual hauls
4) Ratio estimator
variance
(at haul-level)
Non-Groundfish
(sampled hauls)
Groundfish
(sampled hauls)
6) Variance
from percent
retained
Imputation (species, % retained)
(to not-sampled hauls within vessel, gear)
• Data consumers include inseason managers who insure catch
is within specified limits, stock assessment scientists who use
catch information in their assessments, industry managers
who monitor catch allocations and bycatch amounts, agency
analysts, policy makers, and the public.
6
At-sea discard
of non-target species
by haul, species for
observed trips,
sampled hauls
5
5) Imputation
variance
6
6
• The continuing challenge is to implement rigorous
methods while meeting the need for expedited
information for quota monitoring and inseason
management. Evaluation and improvement of
the catch estimation procedures presented here
is an ongoing effort that will result in estimators
that are more robust and continue to meet
management needs.
3) Within and
between haul
sample variance
3
Randomized sample of
hauls and catch on observed trips:
2nd and 3rd levels
of hierarchical sample
• Federal fisheries off Alaska are scientifically monitored and legally enforced
under a number of federal requirements.
• NOAA Fisheries, Alaska Region, assesses catch by integrating
industry-supplied information with data collected by an at-sea and
shoreside observer program administered by the Alaska Fishery
Science Center.
Complete enumeration
of King and Chum
salmon in Bering Sea
pollock fishery
(Amendment 91)
Retained catch
by haul, species
for observed trips
At-sea discard of groundfish
by haul, species
for observed trips
6) Variance
from percent
retained
Post-stratify
all hauls (within strata)
CPs/M full coverage;
CVs all landings
Compute discard rate for each
post-strata (within strata)
Sum estimates across strata,
CP/M observer estimates
and CV landings data
9
Cahalan, J., Mondragon, J., and Gasper, J.
2014. Catch sampling and estimation in
the federal groundfish fisheries off Alaska:
2015 Edition. NOAA Tech. Memo. NMFS-AFSC-286:
42 p. http://www.afsc.noaa.gov/Publications/
AFSC-TM/NOAA-TM-AFSC-286.pdf
Expand to retained
groundfish weight within
post-strata (within strata)
Exclude
between post-strata
(within strata) variance;
accumulate within
post-strata variance
Acknowledgments
Poster by Jason Gasper, Alaska Regional Office, NMFS/NOAA;
Jennifer Cahalan, Pacific States Marine Fisheries Commission;
and Jennifer Mondragon, Alaska Regional Office, NMFS/NOAA;
see NOAA Technical Memorandum AFSC-286 for additional detail.
Exclude between
strata variance;
accumulate within
strata variance
We thank the many groundfish observers who are deployed to commercial fishing
vessels in Alaska to collect high-quality data for science and management,
and the dedicated programmers and application developers who create and maintain
the Catch Accounting System.
7) Variance from
post-stratification
(Number of
observations in each
post-strata varies)
7
8) Ratio estimator
variance
10
10) Variance
from
expansion to
unobserved
trips
Compute discard rate for each
post-strata (within strata)
10
11
Expand to total groundfish
(retained and discarded) weight
within post-strata (within strata)
11
Sum estimates
across post-strata
within strata
11) Sum variance
across
post-strata
12
12
Sum estimates
across strata
Post-stratify
sampled hauls (within strata)
8
8
Sum estimates
across post-strata
within strata
7
11) Sum variance
across strata
Sum domain estimates
across strata
Flow chart adapted from Cahalan et al. 2014. Graphic design by Rebecca White, AFSC.
The recommendations and general content presented in this poster do not
necessarily represent the views or position of the Department of Commerce, the
National Oceanic and Atmospheric Administration, the National Marine Fisheries
Service, or the Pacific States Marine Fisheries Commission.
At-sea Discard
of Groundfish
Retained
Catch
Total Groundfish
Data consumers include inseason managers who insure catch
is within specified limits, stock assessment scientists who
use catch information in their assessments, industry
managers who monitor catch allocations and bycatch
amounts, agency analysts, policy makers, and the public.
At-sea Discard of
Non-Target species