Limits of using the commercial catch data
The use of commercial catch declaration limits the possibility to analyze phenology relative to smaller size classes. In our approach, we filtered the mature component of the populations based on the size distribution of each commercial category as described in Alglave et al. (2023). A similar approach was used in Azevedo and Silva (2020) (though their approach was more refined as proportion by length was considered to vary by size category and by zone also) and allowed to map different age classes of horse mackerel. But a strong limit of our approach is that it is hardly applicable to smaller size classes. Indeed, commercial catch declaration data likely provide a biased picture of the spatio-temporal distribution of smaller fish size class. The minimum landing size is often pretty high and individuals below this size are either rejected or not declared (Lehuta and Vermard, 2023). Furthermore, VMS data are not available for vessels below 12 m that may represent an important part of coastal vessels fishing near juvenile habitats.
As a consequence, mapping juveniles’ habitat is not possible with catch declaration data only. In this case, nursery surveys are still the only available data to investigate juvenile spatio-temporal dynamics and should be considered as reference (e.g. Nurse, Nourseine et Noursom - Delaunay and Brind’Amour, 2018).
Also, another drawback when using commercial data is the targeting behavior of fishermen. This can lead to biased spatial predictions and to overestimated estimates of biomass. Here, we neglected it as trawlers don’t have a strong targeting behavior. Still, the framework developed in Alglave et al. (2022) could be useful in the case of stronger preferential sampling (see Quemper, 2021).