
NuSEDS estimate classification crosswalk
Source:R/nuseds-method-crosswalk.R
nuseds_estimate_classification_crosswalk.RdReturn a static crosswalk of NuSEDS ESTIMATE_CLASSIFICATION values to the
released gcdfo Hyatt (1997) estimate-type concepts
(gcdfo:Type1–gcdfo:Type6, skos:Concepts under gcdfo:EstimateType).
Details
Two families of values deliberately map to no Type concept, and the distinction is recorded here rather than forced:
NO SURVEY THIS YEARis an absence-of-observation marker, not an estimate type – assigning any Hyatt type would assert a survey quality for a survey that did not happen. It maps toNAwith a note.RELATIVE: CONSTANT MULTI-YEAR METHODS/RELATIVE: VARYING MULTI-YEAR METHODSare real classifications with no released concept of their own; they link at scheme level (gcdfo:EstimateType), the same conventionnuseds_estimate_method_crosswalk()uses forCumulative CPUE.
Examples
nuseds_estimate_classification_crosswalk()
#> # A tibble: 10 × 4
#> nuseds_value estimate_type ontology_term notes
#> <chr> <chr> <chr> <chr>
#> 1 TRUE ABUNDANCE (TYPE-1) Type-1 gcdfo:Type1 ""
#> 2 TRUE ABUNDANCE (TYPE-2) Type-2 gcdfo:Type2 ""
#> 3 RELATIVE ABUNDANCE (TYPE-3) Type-3 gcdfo:Type3 ""
#> 4 RELATIVE ABUNDANCE (TYPE-4) Type-4 gcdfo:Type4 ""
#> 5 RELATIVE ABUNDANCE (TYPE-5) Type-5 gcdfo:Type5 ""
#> 6 PRESENCE-ABSENCE (TYPE-6) Type-6 gcdfo:Type6 ""
#> 7 NO SURVEY THIS YEAR NA NA "Abse…
#> 8 RELATIVE: CONSTANT MULTI-YEAR METHODS NA gcdfo:EstimateType "No r…
#> 9 RELATIVE: VARYING MULTI-YEAR METHODS NA gcdfo:EstimateType "No r…
#> 10 UNKNOWN NA NA "Admi…