The NOAA IR serves as an archival repository of NOAA-published products including scientific findings, journal articles, guidelines, recommendations, or other information authored or co-authored by NOAA or funded partners.
As a repository, the NOAA IR retains documents in their original published format to ensure public access to scientific information.
i
Reducing uncertainty in survey abundance estimates by considering alternative designs and estimators: a case study with 3 species in the Gulf of Alaska
-
2023
-
-
Source: Fishery Bulletin, 121(1-2), 50-66
[PDF-1.97 MB]
Details:
-
Journal Title:Fishery Bulletin
-
Personal Author:
-
NOAA Program & Office:
-
Description:A bottom-trawl survey with a stratified-random sampling design has been used to inform stock assessments for commercially important species in the Gulf of Alaska since 1984. A new stratified sampling design was evaluated to determine whether its use could improve the precision and accuracy of abundance estimates. In this proposed approach to defining strata, historical survey data are used to generate what we refer to as information scores (ISes). We compared the traditional stratification scheme with the new method and 2 other sampling designs, using both a design-based estimator and a model-based estimator with each design, to determine if the existing approach is optimal. Statistical robustness, measured in terms of coefficient of variation, bias, and root mean square error, was compared among 7 scenarios with different combinations of estimators and sampling designs by using simulation with a spatiotemporal generalized linear mixed model conditioned on historical observations of catch per unit of effort of 3 species. The combination of the design-based estimator with the IS-based stratification scheme was the best scenario across all performance metrics for all species. This scenario consistently had the lowest variance and smallest total error, and it was generally unbiased. In contrast, the pairing of the model-based estimator with this sampling design was by far the worst-performing scenario. The performance of the existing approach was average.
-
Keywords:
-
Source:Fishery Bulletin, 121(1-2), 50-66
-
DOI:
-
ISSN:0090-0656
-
Format:
-
Publisher:
-
Document Type:
-
License:
-
Rights Information:CC0 Public Domain
-
Compliance:Submitted
-
Main Document Checksum:
-
Download URL:
-
File Type:
