fishing fleets in dat

[sgaichas]

add fishing fleets into datasets
three current fleets include

• demersal trawl
• pelagic trawl
• fixed gear

[gavinfay]

@sgaichas do we have this breakdown for a version of the GB input file for these three fleets?

[sgaichas]

@gavinfay unfortunately not yet. I started to compile data by gear in early April, but the data we had pulled has many unexpected NA values for gear type, and we are having difficulty diagnosing the problem due to a recent requirement to use a new database access package that is giving different results from our current package (no longer allowed on servers). @slucey and @andybeet are working on this

[sgaichas]

@gavinfay I have temporary catch data by gear.

My question is how to specify the q indicator correctly. With catch and discards included, all species are caught in all fleets (though some are in trace amounts).

Should the species with the most catch in the fleet overall be the indicator species for that fleet? Or should each species be associated with the fleet catching the most of that species?

Please see below:

[sgaichas]

The above plot was total catch (landings + estimated discards)

This plot is landings only:

Based on this I would suggest

pelagic trawl (fleet 3) targets Atlantic herring and Atlantic mackerel
fixed (fleet 2) targets spiny dogfish and winter skate
demersal trawl (fleet 1) targets goosefish, silver hake, winter flounder, yellowtail flounder

haddock and cod are my main question: assign them to demersal? can they be targets of both demersal and fixed?

[sgaichas]

A code-able alternative generates the matrix based on the gear with max landings aggregated over the model period. The difference is that only spiny dogfish are assigned to fixed gear as a target:

> fleetdef
# A tibble: 3 × 11
  fishery   Atlantic_cod Atlantic_herring Atlantic_mackerel Goosefish Haddock Silver_hake Spiny_dogfish Winter_flounder Winter_skate Yellowtail_flounder
  <chr>            <dbl>            <dbl>             <dbl>     <dbl>   <dbl>       <dbl>         <dbl>           <dbl>        <dbl>               <dbl>
1 demersal             1                0                 0         1       1           1             0               1            1                   1
2 fixedGear            0                0                 0         0       0           0             1               0            0                   0
3 pelagic              0                1                 1         0       0           0             0               0            0                   0

(code)

# qind supposed to represent target species for fleet
    # define based on majority landings over full time series
    fleetdef <- fishindex %>%
      dplyr::filter(year %in% modyears,
                    variable == "commercial landings") %>%
      dplyr::group_by(Name, fishery) %>%
      dplyr::summarise(totwt = sum(value, na.rm = TRUE)) %>%
      dplyr::select(species=Name, fishery, totwt) %>% 
      dplyr::group_by(species) %>%
      dplyr::slice_max(totwt) %>%
      dplyr::mutate(totwt = 1) %>%
      tidyr::pivot_wider(names_from = "species", values_from = "totwt", values_fill = 0) %>%
      dplyr::arrange(fishery)

[sgaichas]

@gavinfay I pushed new input files to hydra_sim with these fleets and new length bins.
hydra_sim commit 7550767199235862322673eeeaec6b7f54ed07b1

they are in the GB-input folder, filenames hydra_sim_GB_new5bin_1978_3gearF

I am able to run the model with the current hydra_sim.tpl executable.