Source
https://www.canterbury.ac.nz/content/dam/uoc-main-site/documents/pdfs/d-other/geography-repository/GEOG402-13-Local-Food-Initiatives.pdf — original source (opens in a new tab; the file is not redistributed)
UC GEOG402 (2013) — Local Food Initiatives
🔴 The correction source. This document is the reason CR_055 carries
contested: true.The error, verbatim from the source’s own Table 4:
In 192 Beds ......... 98.26 <- SERVINGS per day People per day ......... 33 <- the actual carrying capacityand in text: “a 0.4 acre parcel of land can therefore feed, on average, 33 people per day.” CR_055 reports the 98.26 as PEOPLE per day. It is SERVINGS. A 3× overstatement of vegetable-bed carrying capacity. Had it been applied, the vegetable land requirement would have been understated threefold. ⚠
data_quality: lowdeliberately — a student report, not peer-reviewed; and its yields are not NZ data (Table 1 lbs/acre from Relf (2009) and the Garden of Eden Project (2013); bed geometry from SPIN Farming, US/Canada). A US-yield calculation applied to a Christchurch site.
Summary
A University of Canterbury GEOG402 student report on local food initiatives, 20 pp, examining the vegetable-growing carrying capacity of a Christchurch parcel. It calculates, from bed geometry and per-crop yields, how many servings a 0.4-acre parcel of intensive beds can produce per day, and converts that to how many people it can feed.
Its importance to the corpus is not its analysis, which is derivative and non-NZ in its inputs. It is that this document is the primary that CR_055’s synthesis mis-read, and the mis-read was by a factor of three in the direction that would have made a community’s vegetable land requirement look three times smaller than it is. Reading the primary itself is what turned a plausible-looking figure into a caught error.
Key claims
- claim: "🔴 THE ERROR, verbatim from the source's own Table 4: 'In 192 Beds ......... 98.26' (SERVINGS per day) and 'People per day ......... 33' (the actual carrying capacity); and in text, 'a 0.4 acre parcel of land can therefore feed, on average, 33 people per day.' CR_055 reports the 98.26 as PEOPLE per day. It is SERVINGS — a 3x overstatement of vegetable-bed carrying capacity. Had it been applied, the vegetable land requirement would have been UNDERSTATED THREEFOLD. [verified verbatim against the PDF]"
source_location: "UC GEOG402 (2013), Local Food Initiatives — Table 4 and accompanying text."
- claim: "⚠ WHY data_quality IS low: this is a STUDENT REPORT, not peer-reviewed. Its yields are NOT NZ data — Table 1 lbs/acre come from Relf (2009) and the Garden of Eden Project (2013), and the bed geometry is from SPIN Farming (US/Canada). It is a US-yield calculation applied to a Christchurch site. [verified against the source]"
source_location: "UC GEOG402 (2013), Local Food Initiatives — Table 1 source attributions and methodology."
- claim: "DERIVED CROSS-CHECK: the source's own figures imply approximately 49 m² of parcel per person, which independently brackets OT_190's 53.1 m²/person and its measured 25-65 m²/person household survey range. ⚠ VEGETABLES ONLY — never a whole-diet coefficient. [the ~49 m²/person and the 53.1 m²/person comparison are DERIVATIONS made by the verification pass, not figures printed in either source]"
source_location: "Derived by the verification pass from UC GEOG402 (2013) Table 4 (0.4 acre / 33 people per day) and compared against Millar & Bould (2015) (OT_190); neither m²/person figure is printed in its source."Neobiome Intelligence relevance
Feeds d02_smart_food_systems as a correction and a weak triangulation point — in that order.
1. The correction (the reason this page exists).
| Quantity | CR_055 asserted | The source actually says |
|---|---|---|
| 192 beds, 0.4 acre parcel | 98.26 people/day | 98.26 servings/day |
| People fed | — | 33 people/day |
| Effect if applied | vegetable land requirement 3× too small | — |
This is a units error of exactly the same species as OT_190’s suspected dairy-product-vs-milksolids mismatch, and of the same species as the hazelnut in-shell-vs-kernel block (OT_182) and the milkfat-vs-milksolids trap (OT_172). Four of the most consequential findings across these food-domain sources are unit errors, not data gaps.
2. The triangulation (weak, but real). The source’s own figures imply roughly 49 m²/person of parcel for vegetables. That sits inside the corpus’s other NZ vegetable-space evidence:
| Route | Figure | Source |
|---|---|---|
| Derived from GEOG402’s own carrying capacity | ~49 m²/person | OT_193 (derived) |
| Derived from Blueskin’s land table | 53.1 m²/person | ot_190_brct-2015-blueskin-karitane-foodshed (derived) |
| Measured household survey, residential | 65 m²/person | ot_190_brct-2015-blueskin-karitane-foodshed |
| Measured household survey, large lifestyle blocks | 25 m²/person | ot_190_brct-2015-blueskin-karitane-foodshed |
Three independent routes land in the 25–65 m²/person band. That is genuine convergence and it is worth stating — but note that two of the four figures are derivations, not printed values, and OT_193’s inputs are American. It is a sanity band, not a coefficient.
🔴 Vegetables only — never a whole-diet coefficient. The whole-diet figure is OT_190’s ~0.49 ha/person, of which 81% is meat and dairy. Confusing a vegetable m²/person figure with a whole-diet land requirement would be an error of roughly two orders of magnitude.
No engine cell is auto-changed by this ingest.
Notes
data_quality: lowis deliberate and should not be raised. Student report, not peer-reviewed, with US yields (Relf 2009; Garden of Eden Project 2013) and US/Canadian bed geometry (SPIN Farming) applied to a Christchurch site. It is ingested for what it says, because a synthesis mis-quoted it — not because it is good evidence.- The two m²/person figures on this page are derivations. Neither ~49 (from OT_193) nor 53.1 (from OT_190) is printed in its source; both were computed here. They are recorded as such and must not be re-cited as source figures.
- Vintage: 2013, Christchurch.
Research targets
Documents to retrieve
- (none) — this is the primary; its own upstream yields (Relf 2009, Garden of Eden Project 2013, SPIN Farming) are US material and not worth retrieving for an NZ model.
Research gaps
- Not raised as RTs (recorded here instead):
- A measured NZ vegetable carrying-capacity figure. The 25–65 m²/person band rests on one measured survey (OT_190) plus two derivations. The corpus’s per-area veg yields (CR_021, LIT_058) are UK/synthesis-based, so the whole veg-area chain is thinner than the convergence suggests.
- Why residential sections out-grow large lifestyle blocks per person (65 vs 25 m²/person in OT_190’s survey). A behavioural finding with real design implications for a community’s lot layout; unexplained in either source.
Connections
Links to
Referenced by
EDT domains (1): D02: Smart Food Systems & Agriculture