PacFIN gear codes

[jchampag]

Hey I'm trying to clarify gear codes and gear groups in PacFIN data.
I got access to this document but I am not clear with the last section:

2 DRG ALL DREDGES ALL DREDGE GEAR
2 HKL ALL HOOK&LINE ALL HOOK AND LINE GEAR EXCEPT TROLL
2 MSC ALL OTH GEARS ALL OTHER MISCELLANEOUS GEAR
2 NET ALL NETS ALL NET GEAR EXCEPT TRAWL
2 NTW ALL NON-TRAWL NON-TRAWL GEAR
2 POT ALL POT&TRAP ALL POT AND TRAP GEAR
2 TLS ALL TROLLS ALL TROLL GEAR
2 TWL ALL TRAWLS ALL TRAWLS EXCEPT SHRIMP TRAWLS
2 TWS ALL SH-TRAWLS ALL SHRIMP TRAWLS

Is this just to describe what's each group name stands for ?

Also I have in my data base some samples (more than 700) where PACFIN_GEAR_CODE = 'HKL' or 'TWL' whereas these are categories of PACFIN_GROUP_GEAR_CODE.
I think that just adding them to the HKL and TWL gear groups make sense, is it ok ?

Thanks

[okenk]

Hi Juliette,

PacFIN has a lot of different users and three different state agencies that provide data. As a result, there are a LOT of different columns to denote gear. For stock assessment, we most commonly use PACFIN_GROUP_GEAR_CODE. This is only available for the catch data-- but you can see which specific PACFIN_GEAR_CODE matches up with the more general PACFIN_GROUP_GEAR_CODE. Note that for widow rockfish, bottom trawl and midwater trawl will be combined into the TWL group gear code.

In addition, I found the R code that was used to analyze fishery dependent data in 2019. I have pushed it here.

There is no comparable R code archived for yelloweye I don't think, but @gertsevv has been able to share more detailed spreadsheets than are available in the report with the team.

-Kiva

[okenk]

Also, to answer your specific questions:

1. Yes, it looks like what you have found describes the different PacFIN gear codes.
2. Yes, a PACFIN_GEAR_CODE of HKL should be in the hook and line fleet. TWL is slightly more complicated, as it could belong in the bottom trawl or the midwater trawl fleet. If the year is after ~2000, then TWL should be bottom trawl, because a separate code existed for midwater trawl (MDT).

Note that you can match up the fish ticket in the BDS data (FTID) with the catch data (also FTID) to get the PACFIN_GROUP_GEAR_CODE and also DAHL_GROUNDFISH_CODE (which is necessary to separate out the catches in the shoreside hake fishery).

I will add, when sharing the R code (which I am not sure will be very helpful, unfortunately), I noticed that the 2019 update only updated data for 2015-2018, and retained all other data from the 2015 base model (i.e., did not reanalyze it). I will let other instructors also weigh in, but given that the fleet structure for this model is somewhat complex and older data requires more subjective decisions because it does not match the level of detail of the complex fleet structure, I would do the same and only enter fishery data into the model for 2019-2024. That being said, to make sure you are summarizing the data in the same way, I would still compare what you get prior to 2019, but I would probably not pay too much attention to any differences before ~2000 or so.

[jchampag]

Thanks for your answer !
I should have said on my question, I'm in the yelloweye group.
I feel like cleanPacFIN() is taking care of these HKL and TWL cases I was wondering about.

I am not clear what are the implications of belonging to the bottom trawl or the midwater trawl fleets because I feel like they are analyzed together as a trawl group. Or should each gear been analyzed separately?

[okenk]

Ah, the yelloweye fleet structure is much simpler in terms of gear groupings, for some reason I thought you were working on widow. For yelloweye, the bottom and midwater trawls are in the same fleet, don't worry about that.

Often the gear groupings for TWL and TWS (shrimp trawl) are combined, and then everything else goes into non-trawl. You might check with people analyzing the catch data to make sure that is the case. I have seen shrimp trawl catches put into non-trawl fleets before. Whatever you do should be consistent between the composition data and the landings.

[jchampag]

Ok I will check thanks !

[iantaylor-NOAA]

I support comparing re-analyzed data to what was in the 2019 model but not necessarily replacing the original data, especially if that was done in the previous update. The exception might be that the final year of data in the model often sees the most change because it may not have been complete at the time of the last update. Therefore you may wish to update 2018 as well: 2018-2024.

Regarding gear codes in PacFIN and fleet structure, I've just added some additional R scripts from the 2015 Widow assessment to https://github.com/pfmc-assessments/Assessment_Class/blob/main/Archived%20assessment%20models/Widow%20rockfish/R%20code/2015_Rcode/, where the pacfin.R file has some additional info on how gear codes were separated into midwater vs bottom trawl at that time:
pcatch$gear[pcatch$GRID%in%c("OTW","MDT")] <- "MidwaterTrawl".

[jchampag]

Thanks Ian, I feel like this is more details than I need because I'm working on yelloweye but it would probably be of interest for widow group!