Estimate livestock nitrogen, carbon and volatile-solids excretion.

Converts realised feed intake (the output of redistribute_feed()) into excreted nitrogen, carbon and volatile solids per year x territory x sub_territory x livestock_category. All excretion methods share one canonical nitrogen intake, n_intake = sum(intake_dm_t * feed_n_content), so the methods are directly comparable.

Carbon is the carbon of the excreted organic matter, vs_excretion * c_vs_fraction, so it inherits the intake-and-digestibility mass balance of the volatile solids. Urine carbon is not part of the volatile solids under method_vs = "intake_digestibility" and is not counted: at a C:N near 0.9 it is under a tenth of a dairy cow's excreted carbon (Dijkstra et al. 2018, Table 5) and is respired within days of deposition, so it never reaches the soil carbon this feeds.

Usage

estimate_n_excretion(intake, options = list())

Arguments

intake

A tibble of realised feed intake with at least year, territory, sub_territory, livestock_category, item_cbs_code, feed_quality and intake_dm_t (the redistribute_feed() result). The output of get_feed_intake() does not fit: it is keyed by area_code and live_anim_code and carries no intake_dm_t, and it is refused with a pointer to redistribute_feed(). territory is a stringified area_code (as.character(area_code), what redistribute_feed() emits and what the whole manure chain carries through to the nitrogen inputs); an iso3c literal is still resolved there but is deprecated, since it can only answer with an aggregation bucket and so loses the territory for 62 of the 257 codes it knows.

options

A named list of method options:

method: "intake_minus_retention" (default, n_intake * (1 - n_retention_frac)) or "intake_minus_product_n" (n_intake - product_n).
method_vs: "intake_digestibility" (default, intake_dm_t * (1 - digestibility) * (1 - ash)).
method_digestibility: where digestibility comes from. "species_feed_quality" (default) looks it up per species and feed quality: pigs and poultry fed concentrates digest far more of them than a ruminant does, so they take the IPCC (2019) Table 10.2 (Updated) values for confined monogastrics; see Details. "feed_quality" is the earlier species-blind lookup, one value per feed quality for every species, kept for comparison (whep#1007).
method_c: "volatile_solids" (the only method, vs_excretion * c_vs_fraction). The route it replaced, n_excretion times the bio_coefs Excreta C:N, is deliberately not offered as an alternative: it applied a fresh-dung C:N to whole-excreta nitrogen, some 60% of which is urinary and carries almost no carbon, and returned 0.73 kg C per kg of volatile solids – above pure protein, so not a composition organic matter can have (whep#1006). A route that cannot be right is not a sensitivity case, so this option names the algebra behind c_excretion rather than a choice, and method_c_excretion has one value until a second defensible method exists (whep#1100).
c_vs_fraction: carbon per unit of volatile solids, kg C / kg VS. Default 0.47; see Details.
forage_n: nitrogen content of the grazed forage that intake rows with no item_cbs_code take. "assumed_midrange" (default, 0.02 kg N/kg DM, an assumed unverified value), "gleam_grass_fresh" (0.022), "gleam_grass_hay" (0.017) and "gleam_grass_mean" from GLEAM 3.0 Supplement S1 Tab. S.3.3, or "biomass_coefs_grass" (0.0174) from the bio_coefs Grass row.
product_n: a tibble (year, territory, sub_territory, livestock_category, product_n) required by "intake_minus_product_n".

Value

A tibble with one row per year x territory x sub_territory x livestock_category and columns n_intake, n_excretion, c_excretion, vs_excretion, method_n_excretion, method_vs, method_digestibility, method_c_excretion and method_forage_n.

Details

The default c_vs_fraction of 0.47 kg C per kg of volatile solids is the carbon content of the components that make up faecal organic matter in Dijkstra et al. (2018, Front. Sustain. Food Syst. 2:63, doi:10.3389/fsufs.2018.00063, Table 1): fibre 0.44, microbial organic matter 0.47, protein 0.52 and lipids 0.75 g C per g dry matter, weighted towards the fibre and microbial debris that dominate faeces. Measured manures bracket it: 0.52 for fresh bedded dairy manure (Choi et al. 2022, PeerJ 10:e14134, doi:10.7717/peerj.14134, Table 1: 43.3% C and 83.3% VS of dry matter) and 0.39-0.46 for stored cattle and pig manure (Baek et al. 2020, Int. J. Environ. Res. Public Health 17:4737, doi:10.3390/ijerph17134737, Table 1).

Digestibility

Digestibility here is the digestible share of the feed's gross energy (DE), which is what IPCC (2019) Eq. 10.24 uses for volatile solids. The per-feed-quality values are a scaffold anchored on the GLEAM 3.0 ruminant feed table (gleam_feed_digestibility: fresh grass 66 and hay 58, straws 44-47) and are applied to every species under "feed_quality". Under "species_feed_quality" three of them are replaced by the representative values of the IPCC 2019 Refinement, Vol. 4, Ch. 10, Table 10.2 (Updated), taking the mid-point of each range:

pigs, high_quality: 0.85, growing swine in confinement (80-90). Market swine are the chain's representative pig category; mature confined swine are 70-80. The GLEAM 3.0 pig feed table (gleam_feed_conversion_ratios, Table S.3.4) agrees: maize 88, wheat 83, barley 76, soybean meal 80.
pigs, scavenging: 0.60, free-range swine (50-70).
poultry, high_quality: 0.89, broilers in confinement (85-93). Broilers are the representative poultry category of the whole manure chain; layers in confinement are 70-80.

Every other livestock category and feed quality keeps the scaffold value (rabbits too, although they borrow the swine coefficients elsewhere), and so does crop-residue feed for pigs and poultry: the GLEAM pig table gives straw 49 and crop tops 52, the same as the ruminant straws.

Examples

intake <- tibble::tribble(
  ~year, ~territory, ~sub_territory, ~livestock_category,
  ~item_cbs_code, ~feed_quality, ~intake_dm_t,
  2020L, "203", NA, "Cattle_milk", 2513L, "high_quality", 100,
  2020L, "203", NA, "Cattle_milk", NA, "grass", 500
)
estimate_n_excretion(intake)
#> # A tibble: 1 × 13
#>    year territory sub_territory livestock_category n_intake n_excretion
#>   <int> <chr>     <lgl>         <chr>                 <dbl>       <dbl>
#> 1  2020 203       NA            Cattle_milk            11.9        9.49
#> # ℹ 7 more variables: c_excretion <dbl>, vs_excretion <dbl>,
#> #   method_n_excretion <chr>, method_vs <chr>, method_digestibility <chr>,
#> #   method_c_excretion <chr>, method_forage_n <chr>