Premium
Analyzing Relative Abundance Indices with Log‐Linear Models
Author(s) -
Kimura Daniel K.
Publication year - 1988
Publication title -
north american journal of fisheries management
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.587
H-Index - 72
eISSN - 1548-8675
pISSN - 0275-5947
DOI - 10.1577/1548-8675(1988)008<0175:araiwl>2.3.co;2
Subject(s) - abundance (ecology) , analysis of covariance , relative species abundance , logarithm , statistics , scale (ratio) , environmental science , variance (accounting) , covariance , fishery , mathematics , econometrics , geography , biology , cartography , mathematical analysis , accounting , business
Log‐linear models are proposed for the problem of simultaneously analyzing relative abundance indices available from different survey techniques over different areas and years. I show why analysis on the untransformed (original) scale is inappropriate, and why analysis on the logarithmic scale allows for a generally consistent theory. Models for estimating both annual abundance indices (two‐way analysis of variance), and trends in annual abundance (analysis of covariance) are considered. The methods are illustrated by analyzing data from the Japanese commercial longline fishery for sablefish Anoplopoma fimbria in the gulf of Alaska.