Research topic: commercial catch exploration

[gertsevv]

LANDINGS
Plotted PacFIN landings by major gear group, by state:

Majority of landings are in TW and HKL (FG), which are fleets in 2021 assessments.

TW landings by state:

FG landings by state:

Bycatch within At-sea hake fishery is negligible:

DISCARD
2021 estimate of discard fraction for N model.

Pikitch study show that the historical discard of lingcod was negligible

From Rogers dissertation, based om Pikitch data:

Discard rate from EDCP (mid-1990s) is about 10%.

TW landings vs TW discard length comps:
Recent years:

All years:

Questions:
Where were those small discarded fish before 2002?
Do discard comps include shrimp trawl by any chance?

I ran 2021 model treating TW discard as a separate fleet.

[iantaylor-NOAA]

@gertsevv, thanks for adding all these figures.

Where were those small discarded fish before 2002?

Here's the perspective of the 2021 assessment model (which is a combination of assumed structure and estimated parameters):

• The document says "Prior to 1998, all selected fish are assumed to be retained in the commercial fisheries in
  keeping with comments from those familiar with the history of the commercial fishery at
  the pre-assessment workshop and the Pikitch study in which more than 99% of the observed
  lingcod were retained (pers. comm., J. Wallace, NWFSC)."
• The selectivity figures pasted below (North left and South right) show that the models do include up to 50% of the smallest size bins discarded prior to 1998. However, we also estimate selectivity curves which are steeper and shifted toward larger fish for that period.

It's possible that if we allowed more flexibility in the retention for the early period then the selectivity wouldn't have had to shift away from the small fish because the model could attribute the lack of samples in those bins to discards.

[iantaylor-NOAA]

I forgot to add that for the period 1998 to 2001, the small fish are assumed by the 2021 assessment model to be discard but unobserved because we didn't have a WCGOP program.

[gertsevv]

Thanks Ian. Comparison between discarded and landed comps made me think that it must have been some discards in 1980s, and I was puzzled why Pikitch study did not observe it. We can talk more tomorrow. I ran the model with TW discard set a separate fleet. Here is estimated selex for the discard fleet.

[brianlangseth-NOAA]

This topic is relevant to #197 too and worthwhile to determine whether the bio data suggests net should not go with trawl as it was in 2021.