Columns that particular survey programmes add beyond the NARWC handbook, what each is understood to mean, and how confident that understanding is.
Value
A tibble with columns:
profileShort profile name, for
read_narwc(profile = ).programmeThe survey programme.
columnColumn name as it appears in that programme's files.
meaningWhat the column holds.
roleWhat
narwcrdoes with it:"passthrough"if nothing.confidence"confirmed"or"unconfirmed".
Why this exists
A NARWC extract is not the only shape this data arrives in. Individual survey
programmes carry their own derived columns, and a processed "ready for model"
file may have a dozen of them. They are not in handbook Table 1, so
narwc_schema() does not know about them and read_narwc() would otherwise
drop them without comment.
This registry records what is known about them, per programme. It is expected to grow: CCS is the first entry because it is the one whose files this package was originally written against, and it is an exception rather than a representative case.
Detection suggests, declaration acts
read_narwc() will tell you when a file's extra columns match a known
profile, but it will never apply one because a name matched. That restraint is
deliberate, and Tr_SIGHTING is the reason: it means "sighting made from the
track-line" in CCS files, and there is nothing to stop another programme using
the same name for something else. A column name is not a contract. Acting on a
profile requires naming it — read_narwc(profile = "ccs").
The same caution applies in the other direction. A profile's presence in this
registry means the columns have been identified, not that this package
interprets them. role records which are actually used for anything.
Adding a programme
A new profile needs the programme's name, the columns it adds, and — the part
that is usually missing — what each column actually means, from a data
dictionary rather than from the column name. Entries whose meaning has been
guessed are marked confidence = "unconfirmed" and should stay that way until
someone who ran the survey confirms them.
Examples
narwc_profiles()
#> # A tibble: 8 × 6
#> profile programme column meaning role confidence
#> <chr> <chr> <chr> <chr> <chr> <chr>
#> 1 ccs Center for Coastal Studies, Cape Cod … IS_LAT Aircra… pass… confirmed
#> 2 ccs Center for Coastal Studies, Cape Cod … IS_LO… Aircra… pass… confirmed
#> 3 ccs Center for Coastal Studies, Cape Cod … IS_SP… Specie… pass… unconfirm…
#> 4 ccs Center for Coastal Studies, Cape Cod … Tr_SI… Whethe… pass… confirmed
#> 5 ccs Center for Coastal Studies, Cape Cod … OBSSI… Unknow… pass… unconfirm…
#> 6 ccs Center for Coastal Studies, Cape Cod … Effor… An eff… pass… unconfirm…
#> 7 ccs Center for Coastal Studies, Cape Cod … Date_… Date, … pass… unconfirm…
#> 8 ccs Center for Coastal Studies, Cape Cod … Time_… Time, … pass… unconfirm…
narwc_profiles("ccs")
#> # A tibble: 8 × 6
#> profile programme column meaning role confidence
#> <chr> <chr> <chr> <chr> <chr> <chr>
#> 1 ccs Center for Coastal Studies, Cape Cod … IS_LAT Aircra… pass… confirmed
#> 2 ccs Center for Coastal Studies, Cape Cod … IS_LO… Aircra… pass… confirmed
#> 3 ccs Center for Coastal Studies, Cape Cod … IS_SP… Specie… pass… unconfirm…
#> 4 ccs Center for Coastal Studies, Cape Cod … Tr_SI… Whethe… pass… confirmed
#> 5 ccs Center for Coastal Studies, Cape Cod … OBSSI… Unknow… pass… unconfirm…
#> 6 ccs Center for Coastal Studies, Cape Cod … Effor… An eff… pass… unconfirm…
#> 7 ccs Center for Coastal Studies, Cape Cod … Date_… Date, … pass… unconfirm…
#> 8 ccs Center for Coastal Studies, Cape Cod … Time_… Time, … pass… unconfirm…
# The names a profile would keep
narwc_profiles("ccs")$column
#> [1] "IS_LAT" "IS_LONG" "IS_SPECCODE" "Tr_SIGHTING" "OBSSIGHT"
#> [6] "Effort_Type" "Date_UTC" "Time_UTC"