Source
https://www.dairynz.co.nz/resources/resource-list/new-zealand-dairy-statistics-2023-24/ — original source (opens in a new tab; the file is not redistributed)
DairyNZ / LIC — New Zealand Dairy Statistics 2023-24
The anchor source for the whole community-dairy pathway — and the one that gives kg milkfat per cow directly
The annual DairyNZ/LIC national statistics volume, the authoritative first-hand NZ dairy dataset. National basis: 10,485 herds · 4,701,596 cows · 1,703,404 effective ha · 2.76 cows/ha · 400 kg MS/cow (225 kg milkfat + 176 kg protein) · 1,105 kg MS/ha. Table 4.6 herd-test averages by breed is the fat cell the edible-fat model needs: Jersey 3,669 L/cow, 208.7 kg milkfat, 5.71% fat · HF/J cross 4,765 L, 238.3 kg milkfat, 5.05% · Holstein-Friesian 5,250 L, 236.4 kg milkfat, 4.56% · Ayrshire 4,766 L, 209.1 kg, 4.42%. Because milkfat is published in kilograms per cow, the model needs no milk-density assumption and no g/L conversion — the single largest source of avoidable error in a butter/fat calculation is removed. ⚠ The 2.76 cows/ha has a methodology break (see claims) — use ~2.8 as the commercial stocking benchmark.
Summary
New Zealand Dairy Statistics 2023-24 is the joint DairyNZ / Livestock Improvement Corporation annual compendium of the national dairy herd: herd and cow counts, effective hectares, stocking rate, per-cow and per-hectare production, and herd-test averages resolved by breed. It is a first-hand statistical publication built on LIC herd-testing and DairyNZ farm records, not a synthesis, and it is the newest edition in corpus.
For Neobiome it is the anchor of the community-dairy / edible-fat pathway. Three things it supplies that nothing else in corpus does: (1) kilograms of milkfat per cow per year, by breed, which is the direct input to a butter or cooking-fat calculation without any density or composition assumption; (2) breed liveweights that key straight into the feed-demand table of OT_171; and (3) the commercial stocking-rate benchmark against which a low-input community system can be sized. It supersedes the breed-association booklet raw held alongside it (see Notes).
Key claims
- claim: "NATIONAL BASIS 2023-24: 10,485 herds; 4,701,596 cows; 1,703,404 effective hectares; 2.76 cows/ha; 400 kg milksolids per cow, comprising 225 kg milkfat + 176 kg protein; 1,105 kg MS/ha. [verified verbatim against the source]"
source_location: "New Zealand Dairy Statistics 2023-24, national summary section (herd/cow/effective-hectare counts, stocking rate, per-cow and per-hectare production)."
- claim: "⚠ THE 2.76 cows/ha CARRIES A METHODOLOGY BREAK — it is not real destocking. The publication's own note states that the 2023/24 collection methodology changed, producing verbatim 'larger-than-usual variations in total effective hectares, average farm size and stocking rate.' The prior decade ran 2.81–2.87 cows/ha. Use ~2.8 cows/ha as the commercial stocking benchmark; do not read 2.76 as a real decline. [verified verbatim against the source]"
source_location: "New Zealand Dairy Statistics 2023-24, publication note on the 2023/24 change in collection methodology, accompanying the effective-hectare / stocking-rate series."
- claim: "TABLE 4.6 — HERD-TEST AVERAGES BY BREED (the fat cell). Jersey: 3,669 L/cow, 208.7 kg milkfat, 5.71% fat. Holstein-Friesian x Jersey cross: 4,765 L/cow, 238.3 kg milkfat, 5.05% fat. Holstein-Friesian: 5,250 L/cow, 236.4 kg milkfat, 4.56% fat. Ayrshire: 4,766 L/cow, 209.1 kg milkfat, 4.42% fat. Milkfat is published in KILOGRAMS PER COW, so no milk-density or g/L conversion is required. [verified verbatim against the source]"
source_location: "New Zealand Dairy Statistics 2023-24, Table 4.6 (herd-test averages by breed)."
- claim: "TABLE 4.7 — LIVEWEIGHT BY BREED: Jersey 414 kg, Holstein-Friesian x Jersey 492 kg, Holstein-Friesian 537 kg. These key directly into the feed-requirement table of OT_171. ⚠ The Jersey 414 kg sits BETWEEN OT_171's 400 kg and 425 kg rows — NOT at the 375 kg row the upstream research report selected. [verified verbatim against the source]"
source_location: "New Zealand Dairy Statistics 2023-24, Table 4.7 (liveweight by breed)."
- claim: "BREED SHARE OF THE NATIONAL HERD: Holstein-Friesian x Jersey crossbreed 60.4%, Holstein-Friesian 23.9%, Jersey 7.5%. The crossbreed, not either purebred, is the modal NZ dairy animal. [verified verbatim against the source]"
source_location: "New Zealand Dairy Statistics 2023-24, breed-composition section of the national herd."Neobiome Intelligence relevance
Feeds d02_smart_food_systems as the production side of the community-dairy pathway. With OT_171 supplying feed demand and this source supplying output and liveweight, a per-cow dairy block becomes computable end to end:
| Term | Source | Value |
|---|---|---|
| Milkfat output, Jersey | OT_172 Table 4.6 | 208.7 kg milkfat/cow/yr (5.71% fat, 3,669 L) |
| Milkfat output, Holstein-Friesian | OT_172 Table 4.6 | 236.4 kg milkfat/cow/yr (4.56% fat, 5,250 L) |
| Liveweight, Jersey / HF | OT_172 Table 4.7 | 414 kg / 537 kg |
| Feed demand at that liveweight | ot_171_dairynz-ff-ch4-cow-feed-2021 | look up the matching row, then gross up for utilisation + pasture ME |
| Commercial stocking benchmark | OT_172 (corrected) | ~2.8 cows/ha, not 2.76 |
Three things to carry into any use of these figures:
- The 2.76 is a methodology artefact. The publication itself warns of “larger-than-usual variations” in effective hectares, farm size and stocking rate in 2023/24. The prior decade’s 2.81–2.87 is the real commercial band. A model that reads 2.76 as a trend will encode a decline that did not happen.
- These are commercial twice-a-day figures. They are the ceiling for a community system, not its expectation. The once-a-day counterpart — the production system a house cow actually resembles — is LIT_090, which measures −27% milk / −23% fat against twice-a-day.
- Cross-breed dominance matters for defaults. At 60.4% of the herd the HF/J crossbreed is the modal NZ animal (238.3 kg milkfat, 492 kg liveweight) — a better default than either purebred, even though the Jersey has the highest fat percentage.
No engine cell is auto-changed by this ingest. The dairy pathway has production and feed-demand legs but still has no cost leg — no NZ CapEx or OpEx figure for a community-scale dairy exists in corpus (raised as RT_387 below), and the fat-recovery fraction from milkfat to butter is separately unsourced (OT_173, RT_388).
Notes
- Authoritative NZ industry primary, downloaded directly (not AI-prepared). 63 pp.
data_quality: high. - Supersedes a breed-association secondary that is not in the wiki. A Jersey NZ breed booklet (2023) is a breed-association advocacy publication restating DairyNZ/LIC data. Its table header reads verbatim
(NZ dairy Statistics 2021-22), i.e. it is the same lineage two seasons older and second-hand, and it prints different figures (Jersey 5.57% milkfat / 4.16% protein, against OT_172’s 5.71% / 4.26%). ⚠ Its cow counts are also roughly 2.2× OT_172’s breed totals — a different inclusion rule. Do not mix the two sources. OT_172 is the one to cite. - Milkfat vs milksolids — keep them apart. MS (400 kg/cow) = milkfat + protein. Only the milkfat component (225 kg/cow nationally; 208.7–238.3 kg/cow by breed on herd test) is available for butter or cooking fat. A model that reads milksolids as fat overstates the fat output by roughly 80%.
- Vintage: 2023-24 season. The newest dairy statistics edition in corpus.
Research targets
Documents to retrieve
- RT_387 — NZ CapEx and OpEx for a community-scale dairy (shed, milking plant, vat and cooling, effluent handling; and the annual operating cost of running a handful of house cows). OT_172 and OT_171 give the physical legs of the dairy pathway but neither carries a cost, and the corpus’s existing livestock cost cells (OT_091, CR_044) are sheep-and-beef, not dairy. Blocks the ④f cost line for the dairy pathway.
Research gaps
- Not raised as RTs (recorded here instead): a like-for-like effective-hectare series that removes the 2023/24 methodology break (the ~2.8 prior-decade figure is a workaround, not a restatement); and per-breed feed-conversion efficiency, which would let the model choose a breed on fat-per-tonne-of-DM rather than fat-per-cow. Neither blocks the current pathway.
Connections
Links to
Referenced by