Re-calculate BFISH camera size structure based on abundance at depth

[marcnadon]

There is a bump in the camera size structure for smaller fish, which are also mainly encountered in shallower water. This could be related to the number of size samples coming from different depths.

[marcnadon]

The Bfish camera survey length compositions were recalculating as: 1) Calculate the Number of lengths in each 5cm bins per camera drop within each Year X Depth Strata (so N of lengths/Total Camera drop). This is a proxy for density. 2) Multiply this N_per_drop by the total habitat area in a specific depth strata (proxy for absolute abundance by length bin). 3) Merge the length bins across all depth strata to recreate an overall size structure that takes habitat and abundance into account.

This results in some changes in the overall shape of the size structure: Original size structure: New reconstructed size structure:

[MOshima-PIFSC]

Incorporating the reconstructed length comps into the model did not have much of an effect and the fits did not improve. For now we are going to keep the original length comps in the model.

Original length comps

Reconstructed length comps