Skip to contents

Classify crops into the 2-way SJOS-N safe-and-just space.

Source: R/sjos_n_class.R
classify_sjos_n.Rd

Crosses the ecological boundary side with the nourishment side into the six ordered sjos_levels. Per crop (item_cbs_code), the boundary side is "Exceedance" when the crop-country-year's exceedance_n_t is positive and "Within_boundary" otherwise (the all-zero and missing cases fall to "Within_boundary"). The country's nourishment class (nourish, from normalize_nourishment()) is joined by year and area_code and broadcast to each of its crops. The classification paste(boundary_side, nourish) is one of "Within_boundary Under" ... "Exceedance Over", returned as a factor with all six sjos_levels$level levels. This reproduces Global's 2-way remap (Global/R/sjos_n.r:363) at the per-item_cbs granularity Module 4's footprint needs.

The boundary side reads exceedance_n_t, the crop's signed share of its source cell's overshoot pmax(actual - critical, 0), summed to the country (see build_n_boundary_exceedance()). The cell comparison is made before any crop attribution and the overshoot is not capped at the actual pressure. With negative_critical = "keep", a cell whose critical surplus is negative exceeds even at zero actual surplus, its overshoot is larger than its actual surplus, and its within_boundary_n_t is negative; with "clamp" that cell's allowance is zero instead. A crop whose own surplus is negative inside an exceeding cell receives a negative exceedance share, so it falls on the "Within_boundary" side unless positive shares from other cells of the same country outweigh it.

Usage

classify_sjos_n(exceedance, nourishment, level_col = sjos_class)

Arguments

exceedance

A build_n_boundary_exceedance() output at resolution = "country", keyed by year, area_code, item_cbs_code with the mass terms exceedance_n_t, within_boundary_n_t, actual_n_t.

nourishment

A normalize_nourishment() output carrying year, area_code and the nourish class ("Under" / "Adequate" / "Over"), one row per country-year.

level_col

The unquoted name for the classification column. Defaults to sjos_class.

Value

A tibble keyed by year, area_code, item_cbs_code with the mass terms exceedance_n_t, within_boundary_n_t, actual_n_t, the joined nourish class, the boundary_side and the classification column (a factor over sjos_levels$level, named by level_col).

Examples

classify_sjos_n(
  exceedance = tibble::tribble(
    ~year,
    ~area_code,
    ~item_cbs_code,
    ~exceedance_n_t,
    ~within_boundary_n_t,
    ~actual_n_t,
    2010L, 10L, 2511L, 5, 3, 8,
    2010L, 10L, 2513L, 0, 4, 4
  ),
  nourishment = tibble::tribble(
    ~year, ~area_code, ~nourish,
    2010L, 10L, "Over"
  )
)
#> # A tibble: 2 × 9
#>    year area_code item_cbs_code exceedance_n_t within_boundary_n_t actual_n_t
#>   <int>     <int>         <int>          <dbl>               <dbl>      <dbl>
#> 1  2010        10          2511              5                   3          8
#> 2  2010        10          2513              0                   4          4
#> # ℹ 3 more variables: nourish <chr>, boundary_side <chr>, sjos_class <fct>