Source
https://olivesnz.org.nz/wp-content/uploads/2024/12/2024-Grove-Census-Harvest-Data-Report.pdf — original source (opens in a new tab; the file is not redistributed)
Olives NZ — 2024 Grove Census
The second census edition — and the one that turns the olive figure into a range, not a number
National average 5.1 kg fruit/tree (“a significant decrease from 9.2kg per tree in 2023”), oil yield 14.8% (excluding Bay of Plenty / Waikato). About 310,000 trees nationally; 343 tonnes harvested, “the lowest total since 2015”. 🔴 The averaging basis differs from the 2021 census. Verbatim: “This report includes all trees recorded by census respondents, regardless of whether they were harvested.” The 2021 edition (OT_180) declares no equivalent. The two years may not be like-for-like. 🎯 The model-usable finding is the multi-year pattern: kg/tree swings >2× (10.93 / 9.2 / 5.1) while oil yield % is stable at 13.4–14.8% across five seasons. Oil % is the stable parameter; kg/tree must be carried as a range, never a midpoint.
Summary
The 2024 edition of the Olives NZ grove census, three years after OT_180. It reports a national average fruit yield less than half the 2021 figure, the lowest total harvest since 2015, and — importantly — states its own averaging basis, which the 2021 edition did not.
The two editions together are more valuable than either alone, because their disagreement is diagnostic. The verification pass’s verdict on the 2021-vs-2024 gap: it is season and respondent-base and method, with no evidence of a real structural decline — the national tree count is flat at around 310,000. What survives the disagreement is the oil yield percentage, which barely moves (13.4–14.8% across five seasons) while the fruit yield halves. That asymmetry is the finding the model should encode: extraction efficiency is a property of the fruit and the process; kilograms per tree is a property of the season.
This page carries contested: true against OT_180 for that reason — not because either census is wrong, but because they are not directly comparable.
Key claims
- claim: "NATIONAL AVERAGES, 2024 census: 5.1 kg fruit per tree, described verbatim as 'a significant decrease from 9.2kg per tree in 2023'; oil yield 14.8% (excluding Bay of Plenty and Waikato). [verified verbatim against the source]"
source_location: "Olives NZ 2024 Grove Census — national summary (average kg/tree, prior-year comparison, oil yield %)."
- claim: "NATIONAL SCALE: approximately 310,000 trees nationally; 343 tonnes harvested, described verbatim as 'the lowest total since 2015'. [verified verbatim against the source]"
source_location: "Olives NZ 2024 Grove Census — national tree count and total harvest."
- claim: "🔴 AVERAGING BASIS DIFFERS FROM THE 2021 CENSUS, verbatim: 'This report includes all trees recorded by census respondents, regardless of whether they were harvested.' The 2021 census (OT_180) declares no equivalent basis. The two years may therefore NOT be like-for-like. [verified verbatim against the source]"
source_location: "Olives NZ 2024 Grove Census — methodology / basis-of-reporting note."
- claim: "⚠ SAMPLE LIMITS: 45% response rate (64 of 141), against the 2021 census's 101 respondents. 28% of respondents reported 'No Harvest' (2021: 17%). [verified verbatim against the source]"
source_location: "Olives NZ 2024 Grove Census — response-rate and sample-composition notes."
- claim: "REGIONAL RANK INVERTS AGAINST 2021: Hawke's Bay falls from 16.65 to 5.74 kg/tree, while Central Otago rises from 3.77 to 6.27 kg/tree. [verified verbatim against the source, compared against OT_180]"
source_location: "Olives NZ 2024 Grove Census — regional results table, compared with the Olives NZ 2021 Grove Census regional table."
- claim: "🎯 MULTI-YEAR SERIES — THE MODEL-USABLE FINDING: average kg fruit/tree swings by more than 2x across seasons (10.93 in 2021 / 9.2 in 2023 / 5.1 in 2024) while OIL YIELD % is stable at 13.4-14.8% across five seasons. Oil % is the stable parameter; kg/tree must be carried as a range, never as a midpoint. [verified verbatim against the source and OT_180]"
source_location: "Olives NZ 2024 Grove Census — multi-season oil-yield and kg/tree series; 2021 figure from the Olives NZ 2021 Grove Census."Neobiome Intelligence relevance
Feeds d02_smart_food_systems as the second point of the olive series — the one that establishes the shape of the uncertainty.
| Parameter | Behaviour across seasons | Model treatment |
|---|---|---|
| Oil yield % | Stable: 13.4–14.8% over five seasons | usable as a parameter |
| Fruit kg/tree | Swings >2×: 10.93 → 9.2 → 5.1 | carry as a range, never a midpoint |
| Trees/ha | unpublished in either census (RT_392) | blocks any per-hectare figure |
Oil per tree, as a range: approximately 1.5 kg/tree (2021) to ~0.75 kg/tree (2024). That is the whole of what the corpus can say about olives quantitatively, and it is per tree.
Why the 2021→2024 fall must NOT be read as a decline: the verification pass attributes it to season, respondent base and method simultaneously, and notes the national tree count is flat at ~310,000 — there is no evidence of a real structural decline. The response rate also fell (45% of 141 in 2024 against 101 respondents in 2021) and the “No Harvest” share rose (28% vs 17%), while 2024 explicitly averages over all recorded trees regardless of harvest and 2021 declares no basis. A model that fits a trend to these two points would be fitting a methodology change.
No engine cell is auto-changed by this ingest. Olives remain unmodellable per hectare until RT_392 is filled, and the recovery caveat (RT_391) governs any further pressing assumption on top of the census’s own oil yield.
Notes
contested: trueagainst OT_180 — recorded because the two editions are not like-for-like, not because either is in error. The specific incompatibilities: different response base (64 of 141 vs 101 of 163), different “No Harvest” share (28% vs 17%), and an explicitly different averaging basis stated only in 2024.- The regional inversion is a warning against regional point estimates too. Hawke’s Bay 16.65 → 5.74 and Central Otago 3.77 → 6.27 across three years means even the regional ranking is not stable. Regional modelling of olives needs multi-year means, which the corpus does not have.
- The 14.8% oil yield excludes Bay of Plenty and Waikato — a stated exclusion, so it is not strictly a national figure. It nonetheless falls inside the stable 13.4–14.8% band.
- Industry-association census, self-reported by growers —
data_quality: medium, same as OT_180. - Vintage: 2024 census — the newest olive data in corpus.
Research targets
Documents to retrieve
- (none new) — the planting-density blocker is already raised as RT_392 at OT_180 and applies identically here; it is not duplicated.
Research gaps
- Not raised as RTs (recorded here instead, per the batch’s minimise-new-RTs decision):
- A like-for-like restatement of the 2021 census on the 2024 basis. This is what would make the series comparable. It probably does not exist as a document; the honest treatment is the range plus this caveat.
- The intervening census editions (2022, 2023). The 9.2 kg/tree for 2023 is quoted inside this report; the underlying editions were not retrieved. Retrieving them would firm the range’s shape but not change the conclusion that kg/tree is a range and oil % is a parameter.
- Why the regional ranking inverts (alternate bearing, frost, regional harvest decisions). Not answered by either census.
Connections
Links to
Sources (1): OT_180
Referenced by