User menu
Abundance indices for Katsuwonus pelamis (SKJ) in the Indian Ocean
were derived from the European purse seine CPUE series (2010-2021) for
fishing operations made on floating objects (FOBs). GAMM and GLMM
approach were used to standardize the SKJ catch per floating object set.
The GLMM approach has been applied to compare the outputs when using
an alternative modelling approach and both approaches have been compared
to nominal annual CPUE time series. To account for potential effort creep,
additional explanatory variables have been included in the models. FOB sets
have been classified to non-followed FOBs (i.e., randomly encounter FOBs
for which the purse seiner has no previous information) and followed FOBs
with three distinct classes of tracking buoys: without an echosounder, with a
one-frequency echosounder and with a two-frequency echosounder. Densities
of instrumented buoys at the 1∘ ×1∘-month scale and vessel capacity have also
been included as explanatory variables. The time of the set relative to local
sunrise has been estimated by comparing logbook catch-effort data with VMS
vessel trajectory data and this variable has been integrated in the analysis to
account for changes in fish aggregations around the FOBs over the course of
the day. Results of both GLMMs and GAMMs indicate an initial decrease
in standardized skipjack catch per set over the period 2010-2012 followed
by an increase in standardized skipjack catch per over the period 2012-2021.
This paper represents an update of that presented at the data preparatory
meeting with the following major changes: division of followed buoys into
categories based on echosounder technology, inclusion of Spanish data from
2010-2012 and inclusion in GAMM models of an interaction between year
and follow/echosounder tracking buoy categories
