Question concerning inclusion of unaged fish from Station 2 for kriged biomass apportionment

[brandynlucca]

The Matlab implementation assumes that all fish from Station 1 and Station 2 are "unaged" and "aged", respectively. However, there is a small proportion of fish from Station 2 that remain unaged for one reason or another. Currently, unaged fish from Station 2 are omitted/removed from the dataset entirely. Does it make sense for these unaged fish from Station 2 to be "lumped" together with the unaged fish from Station 1 since apportionment of kriged biomass is, conceptually, divided by "aged vs unaged" rather than "Station 1 vs 2"?

[brandynlucca]

As discussed with PR #215, unaged fish should be dropped in the data loader. This should accompany documentation discussing the caveats of how unaged fish are explicitly handled from different data sets (i.e. Stations 1 and 2). This removal should accompany a console message (e.g. "_x_ unaged samples removed from Station 2 dataset").

[brandynlucca]

After discussing this with FEAT and reviewing the original EchoPro implementation, the updated echopop implementation reflected in PR #241 represents the equivalent workflow as in EchoPro. The unaged fish from Station 2 are retained so they can contribute to the overall length-weight regression fit that is then applied to both aged (Station 2) and unaged (Station 1) fish. These unaged fish are ultimately removed anyways later on in the analysis since apportionment is, in part, a function of age-bins.

This can therefore be closed once PR #241 is merged.