I think this is important for the SOP for transparency in our documents, reproducibility, and also consistency between age comps and other data such as length comps and mean weight-at-age. Copying from an email from Rebecca Haehn:
Question: do we subset the agecomp data by performance, haul type, stratum, or age determination method?
Performance: only the hauls with a positive performance code are used to calculate the design-based estimates (including agecom). A negative performance would be something like: the net came up with a huge hole, a crab pot was caught, the net was hung up on the bottom. The negative indicates that an event occurred that caused the net not to fish properly based on the standards set in the NOAA tech. memo. NMFS-F/SPO 65.
Haul type: The design-based estimates only use data from haul type = 3 (Standard bottom sample (preprogrammed station)). Other stations (such as special project tows and crab retows) are not included in the estimates.
The codebook lists meaning of performance values (pg 75/76) and haultypes (pg 80)
Age determination method: this is uploaded by the age and growth group. The scripts do not filter data by this column.
For the ebs design based estimates, the data is filtered by stratum = 10,20,31,32,41,42,43,50,61,62,82,90 for the plusnw_stratum tables and stratum = 10,20,31,32,41,42,43,50,61,62 for the standard area tables.
This prevents the NBS data from being selected and also excludes potential data from the Russian side and north of the Bering Strait.
Other filters are: survey_definition_id (in race_data.survey_definitions), which filter down data that were collected specifically during the Eastern Bering Sea Crab/Groundfish Bottom Trawl Survey (survey_definition_id = 98). The NBS survey_id =143.
I think this is important for the SOP for transparency in our documents, reproducibility, and also consistency between age comps and other data such as length comps and mean weight-at-age. Copying from an email from Rebecca Haehn:
Question: do we subset the agecomp data by performance, haul type, stratum, or age determination method?
Performance: only the hauls with a positive performance code are used to calculate the design-based estimates (including agecom). A negative performance would be something like: the net came up with a huge hole, a crab pot was caught, the net was hung up on the bottom. The negative indicates that an event occurred that caused the net not to fish properly based on the standards set in the NOAA tech. memo. NMFS-F/SPO 65.
Haul type: The design-based estimates only use data from haul type = 3 (Standard bottom sample (preprogrammed station)). Other stations (such as special project tows and crab retows) are not included in the estimates.
The codebook lists meaning of performance values (pg 75/76) and haultypes (pg 80)
Age determination method: this is uploaded by the age and growth group. The scripts do not filter data by this column.
For the ebs design based estimates, the data is filtered by stratum = 10,20,31,32,41,42,43,50,61,62,82,90 for the plusnw_stratum tables and stratum = 10,20,31,32,41,42,43,50,61,62 for the standard area tables. This prevents the NBS data from being selected and also excludes potential data from the Russian side and north of the Bering Strait.
Other filters are: survey_definition_id (in race_data.survey_definitions), which filter down data that were collected specifically during the Eastern Bering Sea Crab/Groundfish Bottom Trawl Survey (survey_definition_id = 98). The NBS survey_id =143.