Report which areas a present-day-ISO3 population source can reach.
Source:R/population_reach.R
population_source_reach.RdEvery population source WHEP reads is keyed on a present-day ISO3 code: the
gdp-population pin and UN WPP 2024 both are. A WHEP area_code names a
territory in the years it reported, and for a dissolved territory no
present-day ISO3 stands for it, so the source has no row and the area drops
out of every per-capita output (see read_population() and
build_food_supply()).
This reports, for each reporting period of each area code, whether the
vocabulary in iso3_codes reaches it:
"direct"— the period's own ISO3 is in the vocabulary."successor"— it is not, but thesuccessorrelation the polities database publishes leads, transitively, to ISO3 codes that are. Those codes are returned iniso3_reached."unreachable"— neither. No arrangement of the source's rows supplies this area.
Measured against UN WPP 2024's 237 Country/Area ISO3 codes, 14 reporting
periods are not "direct" and five are "unreachable" — ATF, PCN,
SGS, NFK and CXR, every one of them inside the Rest-of-World bucket
999. ANT-1961-2010, area 151 Netherlands Antilles, was the last
"unreachable" outside that bucket (#787); the 2026-08-25 upstream re-sync
gave it CUW-2010-2025; SXM-2010-2025; BES-2010-2025, so it now reads
"successor" by lookup rather than by anything hardcoded here.
"successor" says the ISO3 codes exist, not that summing them is a safe
denominator. In general it is not: a successor sum for the Yugoslav SFR falls
17.5% short of the gdp-population pin's own figure for the same aggregate,
of which Kosovo is 42% and a level disagreement between the pin and WPP on
that series is the rest. Use this to see what a source can cover, and read
the note at the top of R/population_reach.R before turning any of it into a
population.
extent_exceeds_iso3 marks the periods where the codes reported are
narrower than the territory, because the walk stopped on a polity that
reuses one of its own parts' ISO3 — SRB-2006-2008 "Serbia (including
Kosovo)" carrying SRB, SUD-1956-2011 carrying SDN (#863). It is
orthogonal to reach: a "direct" period can be flagged, and that is the
case worth knowing about, since nothing else about the row hints at it.
Rows on area_code 999 describe the members of the Rest-of-World fold
bucket rather than the bucket itself, since each member is a crosswalk period
of its own.
Arguments
- iso3_codes
Character vector of the present-day ISO3 codes the population source publishes, e.g.
unique(read_wpp_population()$iso3c).- crosswalk
Optional crosswalk overriding polity_area_crosswalk, for testing.
Value
A tibble with one row per area_code and polity_code reporting
period, sorted by area code then first reported year: area_code,
polity_code, polity_name, map_year_start, map_year_end,
own_iso3 (the period's own ISO3), reach ("direct" / "successor" /
"unreachable"), n_iso3, iso3_reached (the ISO3 codes standing in
for the area, NA when unreachable) and extent_exceeds_iso3 (TRUE
where those codes cover less ground than the period's territory).
Examples
population_source_reach(c("BEL", "LUX", "CZE", "SVK", "CUW"))
#> # A tibble: 297 × 10
#> area_code polity_code polity_name own_iso3 map_year_start map_year_end reach
#> <int> <chr> <chr> <chr> <int> <int> <chr>
#> 1 1 ARM-1991-20… Armenia ARM 1992 2024 unre…
#> 2 2 AFG-1919-20… Afghanistan AFG 1961 2024 unre…
#> 3 3 ALB-1913-20… Albania (1… ALB 1961 2024 unre…
#> 4 4 DZA-1919-19… Algeria (1… DZA 1961 1961 unre…
#> 5 4 DZA-1962-20… Algeria (1… DZA 1962 2024 unre…
#> 6 7 ANG-1905-19… Angola (19… AGO 1961 1974 unre…
#> 7 7 AGO-1975-20… Angola (in… AGO 1975 2024 unre…
#> 8 8 ATG-1800-20… Antigua an… ATG 1961 2024 unre…
#> 9 9 ARG-1902-20… Argentina ARG 1961 2024 unre…
#> 10 10 AUS-1901-20… Australia AUS 1961 2024 unre…
#> # ℹ 287 more rows
#> # ℹ 3 more variables: n_iso3 <int>, iso3_reached <chr>,
#> # extent_exceeds_iso3 <lgl>