Bait

[pwoodworth-jefcoats]

Examine how bait changes over time and space to get a sense of how we can best account for these changes in the BRTs. Do all boats generally do the same thing and change between baits at the same time? Is there a need to account for bait in our gridding data, or can it be a monthly factor?

[pwoodworth-jefcoats]

I took a quick look at bait codes at an annual resolution. Through 2020, four bait codes are used across over 93% of the sets:

• saury/sanma
• sardine
• sanma and sardine
• milkfish

From 2021 through 2024, most of the sets that weren't using these codes were using either "other" or "mixed". You can see a rough heat map in our shared drive

Next up is looking at how bait varies at the monthly scale and over space. I think it's something we need to account for and I'm still hoping we can do it as a factor.

[pwoodworth-jefcoats]

Next step: Create a monthly time series for a bait factor, with bait either being:

• saury/sanma | sardine | sanma and sardine OR
• milkfish | mixed | other

Note the percentage of sets captured by the factor.

[pwoodworth-jefcoats]

Brief update on creating a factor for bait type in our shared drive. Short answer: this should be doable.

[pwoodworth-jefcoats]

I updated the figures I made for determining a bait factor and included the dates for each factor. I aligned the breakpoints with calendar years since they matched so closely. This way, there's the same number of each month for each factor. Not sure whether this matters. We can also adjust the break points, too.

[pwoodworth-jefcoats]

I realized that I didn't restrict the data to our geographic area of interest. This has been corrected and the figures updated. It doesn't change much.

[JohannaWren]

I can't remember if we decided on having just two categories for bait or three, so I'm going with the three factors Phoebe proposed: Sanma/sardine for everything before 2021, transition Jan 2021-Dec 2022, then milkfish starting in Jan 2023. Please let me know if you do not agree with this, or if you think we should have only two categories (and if so, when to split them).