European purse seine CPUE standardization: methodology and framework for the YFT stock assessment

Purse seine CPUE standardization is thought on a combination of fishing mode and commercial size categories
of species basis, i.e., large fish in free schools (FSC) sets on one side and small fish under floating objects
associated sets on the other side. However, while FSC sets are randomly encountered, FOB sets can either be
randomly encountered, e.g., foreign drifting fish aggregating devices (dFADs) or natural log not instrumented, or
not randomly encountered, i.e., vessels have access to buoys and/or echosounder data equipping the dFAD. The
non-randomness of encounter leads to different statistical approaches and different impact on effort creep. On
one hand, the standardization approach using an extension of the Delta-lognormal GLMM to three components,
i.e., the product of the number of schools detected (summing positive and null sets) (number of schools) by spatiotemporal
strata, the proportion of positive sets with the species/category of interest and the catch per positive set
with it (school size), is appropriate to randomly encountered schools. We propose to apply this methodology to
FSC sets as well as to FOB sets randomly encountered. On the other hand, for FOB sets not randomly encountered,
we propose to use, as classical approaches, the product of the third component, i.e., school size, by a fishing
efficiency rate per set calculated with a methodology quantifying the increase in fishing efficiency due to the use
of FOB equipped with echosounders (Wain et al. 2020). This framework would allow to homogenized
standardization of CPUE based on fisheries-dependant data and provide several time series, i.e., on randomly
encountered FSC (> and < 10kg) and FOB sets separately and on not randomly encountered FOB sets, here of
EU purse seine fleet catches per unit effort (CPUE) of yellowfin tuna (YFT) from the Indian Ocean.