Fraction of manure managed in each manure-management system (MMS) by region and species, in two selectable halves: the GLEAM 2.0 ingest that the manure engines read by default, and the unsourced placeholder it replaced.
Format
A tibble with source, region, species, mms_type, fraction
and reference. source is either "gleam_2_0" or "placeholder", and is
the value the manure engines' MMS-shares option takes (see
manure_engine_options). reference cites each row. fraction sums to
one within every (source, region, species). region is the IPCC label
vocabulary .add_ipcc_region() emits, plus a "Global" fallback.
Source
source == "gleam_2_0" is ingested from Tables 4.2-4.11 of Supplement S1
to FAO. 2018. GLEAM Model description, Version 2.0, Revision 5, July
2018, data reference year 2010 (workbook
https://www.fao.org/fileadmin/user_upload/gleam/docs/GLEAM_2.0_Supplement_S1.xlsx,
219715 bytes, md5 72fd2ea477dfe8b30cd3657b2baa4af1, re-downloaded from
FAO and verified 2026-09-09; FAO issues no DOI). Those ten tables are the
only published regional MMS shares in any GLEAM release: version 3.0
dropped the family, and its Supplement S1 carries MMS definitions only.
The read and the crosswalk are in data-raw/livestock_coefficients.R.
The published grain is finer than this table's on three axes at once, so the ingest is a crosswalk and each of its four choices moves numbers (whep#958). Stated here, and at the point of use in the builder:
Species collapse. GLEAM publishes one table per production system; this table is keyed on
species_gen. Systems are averaged unweighted within a species and region, counting only the systems the source publishes a column for. A herd- or manure-weighted mean would be better and is not available from the source: Supplement S1 publishes herd parameters (Tab. 2.4-2.21) and no regional animal numbers or system shares anywhere. Tab. 4.4 (feedlot cattle) is not read: GLEAM models the feedlot as a sub-system of the beef herd, so reading it would give a minority system a third of all cattle manure. The equal-weight collapse is coarsest for poultry, where the flat backyard assumption (50 percent pasture, 50 percent daily spread, identical in all ten regions) gets a third of the weight.Region collapse. GLEAM's 10 regions are mapped onto the IPCC labels
.add_ipcc_region()emits. Only one pair collides – the Russian Federation and Eastern Europe both map to"Eastern Europe"– and those two published columns are averaged unweighted.The
Globalrow. GLEAM publishes no global row, and most WHEP output resolves to"Global"(whep#678), so the row carrying the most weight is the one with no published value. It is the unweighted mean over the 10 GLEAM regions of the species-collapsed distributions, counting a region only where the source publishes one, and taken over the 10 published regions rather than the 9 IPCC labels so that the Russian Federation and Eastern Europe keep one vote each. This row is WHEP-derived, not FAO-published, and itsreferencesays so.MMS vocabulary. GLEAM names 14 systems; WHEP's manure chain serves six labels and only six, in all four tables it reads by MMS name. Pasture, daily spread, solid storage, liquid slurry, uncovered anaerobic lagoon and poultry manure with litter map by identity. Drylot maps to
"Solid Storage", following IPCC's own combined "solid storage and dry lot" category, which ipcc_2019_n2o_ef_direct carries at the same EF3; a separate"Dry Lot"label would give a lower methane conversion factor (2.5 against 4.0 percent, Temperate). Pit storage maps to"Poultry Manure"for chickens (a layer deep pit is IPCC's poultry manure without litter, held at the same EF3 as the with-litter row) and to"Liquid/Slurry"otherwise – assumed, unverified, as is intensive-windrow composting to"Solid Storage". Burned for fuel and anaerobic digester are excluded and the rest renormalised: neither has a servable WHEP label, both being in climate_mcf only at the unjoinableclimate_zone == "All"and inmanure_loss_fractions.csvnot at all. Excluding them says the excreted nitrogen passes through the systems that remain, which overstates those systems; it removes 18.5 percent of published buffalo shares (dung burnt for fuel), 5.2 percent of swine and 4.1 percent of cattle, and none of the others.
GLEAM 2.0 publishes no manure-management table for horses, camels or
mules and asses. Those four rows are the placeholder's, retained so the
engine still resolves a split for them, and flagged unsourced in
reference.
source == "placeholder" is the table WHEP shipped before whep#958:
a hardcoded tribble() annotated "GLEAM 3.0 / FAO statistics
(simplified)", round to the nearest 5 percentage points, matching no
table of any GLEAM release (whep#921). It is kept selectable so the
values published before the ingest stay reproducible and the sensitivity
to the ingest stays measurable. It is not a defensible alternative
estimate.
This object is result-affecting through .resolve_mms_shares(): it
weights the Tier 2 manure CH4 methane conversion factor (climate_mcf)
and the manure direct-N2O emission factor (ipcc_2019_n2o_ef_direct) at
both tiers.
Examples
regional_mms_distribution
#> # A tibble: 231 × 6
#> source region species mms_type fraction reference
#> <chr> <chr> <chr> <chr> <dbl> <chr>
#> 1 gleam_2_0 Asia Buffalo Daily Spread 0.0942 GLEAM 2.0 Su…
#> 2 gleam_2_0 Asia Buffalo Liquid/Slurry 0.391 GLEAM 2.0 Su…
#> 3 gleam_2_0 Asia Buffalo Pasture/Range/Paddock 0.141 GLEAM 2.0 Su…
#> 4 gleam_2_0 Asia Buffalo Solid Storage 0.373 GLEAM 2.0 Su…
#> 5 gleam_2_0 Eastern Europe Buffalo Daily Spread 0.195 GLEAM 2.0 Su…
#> 6 gleam_2_0 Eastern Europe Buffalo Liquid/Slurry 0.0540 GLEAM 2.0 Su…
#> 7 gleam_2_0 Eastern Europe Buffalo Pasture/Range/Paddock 0.207 GLEAM 2.0 Su…
#> 8 gleam_2_0 Eastern Europe Buffalo Solid Storage 0.543 GLEAM 2.0 Su…
#> 9 gleam_2_0 Global Buffalo Daily Spread 0.216 GLEAM 2.0 Su…
#> 10 gleam_2_0 Global Buffalo Liquid/Slurry 0.214 GLEAM 2.0 Su…
#> # ℹ 221 more rows