carcass estimates - confirm run and adipose clipped

[alvizek]

For the carcass estimates data, we need to confirm the run and the adipose status for the data we have (prioritize the data that we think we will use for JPE, Feather and Butte).

[liz-fw]

Butte notes:

• carcass survey is performed for spring run (https://www.calfish.org/ProgramsData/ConservationandManagement/CentralValleyMonitoring/SacramentoValleyTributaryMonitoring/ButteCreek.aspx) so assume that all are spring.
• two types of files for each year 2014-2020 - individuals and chops. Individuals record adipose_fin_clip and so do chops.
• historical estimates in an excel have the results from modeling for 2001-2020, which fill in earlier years but do not have any associated adipose_fin_clipped data. These are not included in standard_carcass.
• individual data from 2019 does not have dates in the raw file.

[liz-fw]

Feather notes:

• Data collected from 2000-2020
• According to notes, run was collected for all years but is only present in 2008-2009 and 2017-2021
• According to notes, adipose_fin_clipped was collected for all years. Due to misalignment of header tables and data tables, there are many NA dates associated with adipose_fin_clipped in the final file and no way of recovering these easily - they get filtered out in the feather_carcass_qc.Rmd
• Reviewed by year. 2000-2001 does not collect either variable. Run not collected for 2000-2007 and 2010-2016
• Fixed joining of cwt data with cwt header (date) for 2008, 2002, 2003, 2004 (still some missing dates due to misaligned header IDs), 2005 (some missing dates same as 2004), 2006 (some dates missing), 2007 (some dates missing)
• Some dates missing in 2009 because of missing some matching header IDs in chopChannel with ChopChannel_header

[liz-fw]

in branch carcass-dat-achecks:

• ensured all available adipose_fin_clip and run data were incorporated into feather_carcass
• added run and adipose_fin_clip to Butte carcass where available - mostly going off reports
• kept 2019 Butte data in the final standard dataset by mutating a fake date (2019-01-01) for those data - confirm this is what we want to do, otherwise they get filtered out

[liz-fw]

Battle Creek: spring run and unknown covered by survey but not recorded in data, so marked run as NA. adipose_fin_clipped recorded.

Clear Creek: data were obtained from spring-run chinook salmon surveys, but contains column run and run_call which have several different encodings (fall, late-fall, spring, hybrid).

Yuba River: two tables (individuals and chops). Individual table has adipose_fin_clip but no run information.

[liz-fw]

associated branch: carcass-dat-achecks (typo is mine, oops)

[liz-fw]

Action items:

• [x] Reach out to Grant and confirm 2019 dates
• [x] Does 2019 include both pre-spawn and post-spawn surveys (we don't want to sum all if so)
• [x] Confirm with Keith - CJS run, adipose_fin_clip?
• [x] Battle: change run to spring
• [x] Clear: use run column instead of run_call
• [x] Yuba: run = unknown

[liz-fw]

e-mails sent to Grant and Keith 11-28-2023. Grant said he will look into those (Butte). Casey responded for Feather and said:

We consider the CJS estimates as a combined spring- and fall-run estimate.

CDFW’s Ocean Salmon Program does a big synthesis on coded wire tag recoveries in the Central Valley annually. As part of that synthesis, they use our sample data and coded wire tag recoveries to expand and estimate the hatchery and natural origin proportions that can then be applied to our estimate. They are usually a few years behind, but I’ve attached their most recent reports so you can see how the expansions are done. You can find all of their reports on the CalFish website here: https://www.calfish.org/ProgramsData/ConservationandManagement/CentralValleyMonitoring/CentralValleyCFMProgram.aspx

[liz-fw]

• [x] need to follow up with Grant on e-mail chain to ask about the CJS confidence intervals.
• [x] add in 2019 data file Grant sent
• [x] push up to cloud (with 2019 data)
• [x] re-run standard carcass Rmd