Skip to contents

Calculate cohort and production system distribution.

Source: R/livestock_cohorts.R
calculate_cohorts_systems.Rd

Distributes national herd totals across GLEAM-defined cohorts and production systems using gleam_livestock_categories and regional weight data.

GLEAM supplies only the taxonomy here: which systems and cohorts each species has. It supplies no herd shares. gleam_livestock_categories has no share column, and the herd-parameter tables of the GLEAM 2.0 and 3.0 supplements give demographic rates and live weights, not a dairy/meat or layer/broiler split of the herd (whep#1194).

Where FAOSTAT already reports the herd split into the items build_primary_production() carries ("Cattle, dairy" / "Cattle, non-dairy", "Pigs" / "Hogs" for market / breeding swine, and "Chickens, layers" / "Chickens, broilers"), the whole herd goes to the system its item names and no share is applied. Every other herd (buffalo, sheep, goats, ducks, turkeys, geese, and any aggregate cattle, swine or poultry label) is split by WHEP's default shares, which are assumed, unverified placeholders with no source. The method_system_share column says which of the two applied to each row.

Within a system, method_cohorts says how the heads reach cohorts. A "Cattle, dairy" head is a cow producing milk: FAOSTAT's dairy cattle are the Milk Animals element of cow milk, and its non-dairy cattle are the rest of the herd, "cattle minus dairy cattle" (Tubiello et al. 2015, Estimating Greenhouse Gas Emissions in Agriculture, FAO, ISBN 978-92-5-108674-2, pp. 45-46). So by default those heads all go to the dairy system's "Adult Female" cohort, and the dairy herd's bulls, heifers and calves are counted where FAOSTAT counts them, in the non-dairy herd (whep#1127). Every other herd is split equally across its system's cohorts, which is assumed, unverified: GLEAM supplies no cohort shares. The method_cohort_share column says which applied to each row.

Usage

calculate_cohorts_systems(
  data,
  system_shares = NULL,
  method_cohorts = c("milk_animals", "uniform")
)

Arguments

data

Dataframe with species, heads, and optionally iso3 or region.

system_shares

Optional dataframe with species_gen, system, system_share columns. If NULL, a herd whose commodity names a production system goes wholly to it, and every other herd uses WHEP's assumed, unverified default shares. Supplying this overrides both, so the supplied shares are used verbatim.

method_cohorts

How heads are split across the cohorts of a system:

  • "milk_animals" (default): a dairy-cattle herd ("Cattle, dairy", the FAOSTAT Milk Animals count) goes wholly to the dairy system's "Adult Female" cohort; every other herd is split equally across its system's cohorts.

  • "uniform": every herd, dairy cattle included, is split equally across its system's cohorts, the behaviour before whep#1127. It books five sixths of the milking cows as bulls, heifers and calves, and is kept for sensitivity analysis.

Value

Dataframe expanded to cohort level with cohort, system, cohort_heads, and cohort_fraction columns, plus method_system_share: "reported" when the commodity itself names the system, "assumed" when WHEP's unsourced default split was applied, or "supplied" when system_shares was given, and method_cohort_share: "milk_animals" for a dairy-cattle herd routed to its milking cows, "uniform" for an equal split across cohorts, or NA for a species with no cohorts. A milk_yield_kg_day column (per head of the input row) is moved onto the milked cohort, the "Dairy" system's "Adult Female", at the yield that keeps the herd's milk unchanged; every other cohort gets 0. method_milk_yield says which: "milked_cohort", "not_milked_cohort", or "whole_herd" for a species with no cohorts. A herd with milk but no milked cohort aborts (class whep_milk_without_milked_cohort).

Examples

tibble::tibble(
  species = "Cattle", heads = 10000,
  iso3 = "DEU"
) |>
  calculate_cohorts_systems()
#> # A tibble: 11 × 10
#>    species heads iso3  species_gen system method_system_share cohort            
#>    <chr>   <dbl> <chr> <chr>       <chr>  <chr>               <chr>             
#>  1 Cattle  10000 DEU   Cattle      Dairy  assumed             Adult Female      
#>  2 Cattle  10000 DEU   Cattle      Dairy  assumed             Adult Male        
#>  3 Cattle  10000 DEU   Cattle      Dairy  assumed             Replacement Female
#>  4 Cattle  10000 DEU   Cattle      Dairy  assumed             Replacement Male  
#>  5 Cattle  10000 DEU   Cattle      Dairy  assumed             Surplus Female    
#>  6 Cattle  10000 DEU   Cattle      Dairy  assumed             Surplus Male      
#>  7 Cattle  10000 DEU   Cattle      Beef   assumed             Adult Female      
#>  8 Cattle  10000 DEU   Cattle      Beef   assumed             Adult Male        
#>  9 Cattle  10000 DEU   Cattle      Beef   assumed             Replacement Female
#> 10 Cattle  10000 DEU   Cattle      Beef   assumed             Replacement Male  
#> 11 Cattle  10000 DEU   Cattle      Beef   assumed             Fattening         
#> # ℹ 3 more variables: method_cohort_share <chr>, cohort_fraction <dbl>,
#> #   cohort_heads <dbl>