RD_029: NZLRI South Island Land Use Capability Extended Legend, 2nd ed. (van Berkel ed., NWASCO 1983/84)

RD_029: NZLRI South Island Land Use Capability Extended Legend, 2nd ed. (van Berkel ed., NWASCO 1983/84)

Source

https://www.mpi.govt.nz/dmsdocument/32359 — original source (opens in a new tab; the file is not redistributed)

NZLRI South Island Land Use Capability Extended Legend (ED 2)

The primary per-LUC-UNIT stock-carrying-capacity source behind the food module's su_ha[LUC] layer (④f) — advances RT_281 (South Island portion)

The South Island volume of the New Zealand Land Resource Inventory (NZLRI) extended legends — the printed reference that OT_083 names as the only place the per-unit, per-region stock-carrying capacities live (they are not NZLRI GIS attributes). For each of the 219 South Island LUC units it tabulates carrying capacity in stock units per hectare (ewe equivalents, annual) at three management levels — Present Average / Top Farmer / Attainable Physical Potential — for DRY and irrigated (IRR) land, across up to four South Island MAF regions (Nelson [incl. Marlborough & Westland], Canterbury, Otago, Southland). Government primary, read direct → data_quality verified (with the OCR caveat in Notes). Editor P. van Berkel, National Water & Soil Conservation Organisation (NWASCO) / Water & Soil Division, Ministry of Works & Development.

Summary

The South Island Extended Legend is the regional companion volume to the NZLRI worksheets: it gives a full one-page technical description of every LUC unit mapped on the South Island 1:63,360 NZLRI sheets — geographic distribution, altitude zone, slope, rock type, soil groups, erosion, vegetation, present/potential land use, and (the NI-load-bearing part) the productivity indices: stock-carrying capacity in SU/ha and Pinus radiata site index. This is the document RT_281 sought: it lets the food model replace its single national per-LUC-class stocking approximation (su_ha[LUC], calibrated only at LUC 6–7) with region- and unit-specific Present-Average carrying capacities for the four South Island MAF regions. This ingest registers and characterises the source and captures the method plus anchor figures verbatim; the bulk extraction of the full 219-unit table into the engine lookup is a downstream data task (see Notes). Edition 2 draft, January 1983 (revised 27.1.84).

Key claims

- claim: "The South Island Extended Legend contains summary details of the 219 land use capability (LUC) units mapped in the South Island part of the New Zealand Land Resource Inventory (NZLRI). The NZLRI was produced by the Water and Soil Division of the Ministry of Works and Development for the National Water and Soil Conservation Organisation. Land use capability is 'an assessment of the land's capacity for sustained productive use taking into account physical limitations, including climate, soil conservation needs and management requirements', classified as Class → Subclass → Unit; there are eight classes, limitation-to-use increasing from Class 1 to Class 8. Edition 2 (draft), editor P. van Berkel, January 1983 (rev. 27.1.84)."
  source_location: "Title page (p.1); §1 Introduction (p.4)"
- claim: "Stock carrying capacity method (verbatim): 'Estimates of the stock carrying capacity at various management levels have been made in collaboration with MAF advisory officers. These are given as the number of stock units per hectare (expressed as ewe equivalents) that are capable of being carried on an annual basis. Figures are given assuming no irrigation (DRY), and with irrigation (IRR) where available and appropriate.' The three management levels are Present Average, Top Farmer, and Attainable Physical Potential. 'The figures are presented for up to 4 regions, corresponding to the MAF regions: Nelson (including Nelson, Marlborough and Westland), Canterbury, Otago and Southland.'"
  source_location: "§ 'Stock carrying capacity' explanatory note (p.8)"
- claim: "Worked anchor — Unit 1c1 (well-developed deep soils in summer moisture-deficient districts; Canterbury Plains + minor Wairau Plain; lowland; 6,460 ha = 0.04% of S.I.; present land use intensive grazing + cereal/forage/horticultural cropping). Stock carrying capacities (SU/ha): PRESENT AVERAGE — Canterbury 13 DRY / 16 IRR, Nelson 10 DRY / 15 IRR; TOP FARMER — Canterbury 15 DRY / 20 IRR, Nelson 15 DRY / 20 IRR; ATTAINABLE PHYSICAL POTENTIAL — Canterbury 18 DRY / 24 IRR, Nelson 20 DRY / 25 IRR. (These same six Canterbury/Nelson figures appear identically in the worked Example §2, p.9, and in the detailed 1c1 entry — double-confirmed.)"
  source_location: "§2 Example, p.9; Unit 1c1 detailed entry, p.11"
- claim: "Low-capability contrast — Unit 5c1 (strongly rolling to moderately steep limestone/volcanic hills with rock outcrops, seasonally moisture-deficient eastern districts; North Canterbury; lowland–lower montane; 2,100 ha = 0.01% of S.I.; present + potential land use = grazing / intensive grazing). Stock carrying capacity: PRESENT AVERAGE Canterbury 6 SU/ha DRY (IRR blank). Illustrates the >2× within-'prime'-classes spread a flat national per-class figure cannot capture (1c1 lowland cropping plain 13 vs 5c1 hill 6 SU/ha, both Canterbury Present-Average DRY)."
  source_location: "Unit 5c1 detailed entry, p.101"
- claim: "South Island LUC class-area distribution (Figure 1, 'Uses of classes 1 to 8'): Class 1 — 36,000 ha (0.27%); Class 2 — 490,000 ha (3.27%); Class 3 — 1,409,000 ha (9.37%); Class 4 — 1,462,000 ha (9.77%); Class 5 — 117,000 ha (0.87%); Class 6 — 3,346,000 ha (22.17%); Class 7 — 2,877,000 ha (19.07%); Class 8 — 4,888,000 ha (32.37%); Rivers, Lakes, etc — 505,000 ha (3.37%). Classes 6–8 (steep/mountain, low capability) = ~73.6% of the South Island; the arable/versatile classes 1–3 = ~12.9%. (⚠ OCR caveat applies to ALL NINE printed percentages — not just Class 1/5: every area/percentage is OCR-read from a 1983 scan and must be treated as approximate, not digit-exact. The three dominant classes 6/7/8 recompute close to the printed values against a ~15.1 M ha S.I. total; Class 1 [0.27% printed vs ~0.24% recomputed] and Class 5 [0.87% vs ~0.77%] show the largest discrepancies — but none of the nine is verified exact. See Notes.)"
  source_location: "Figure 1, p.5"

Neobiome Intelligence relevance

This is the primary source RT_281 was raised to retrieve — the NZLRI regional extended legend for the South Island. It advances (partially resolves) the food-module precision target ④f. The model today differentiates grazing by LUC land class with a single national per-class Present-Average table su_ha[LUC] (model_design.md §4/§5), hard-anchored only at LUC 6–7 (≈7.5 SU/ha, OT_082) with the other classes OT_083-band indicative. This legend supplies the region- and unit-specific Present-Average SU/ha for the four South Island MAF regions (Nelson incl. Marlborough/Westland · Canterbury · Otago · Southland) that would replace the approximation for South Island sites — the model uses Present Average (the realistic level, not the Attainable ceiling), which this document tabulates per unit per region.

It is the empirical counterpart to OT_083 (the methodology handbook): OT_083 says the per-unit carrying capacities “are held in the regional extended legends” and are not in the GIS attributes — this is one of those legends. It confirms OT_083’s hard scope caveats: SU/ha are sheep-and-beef pastoral only (do not apply to dairy or cropping), annualised (ewe-equivalents carried year-round), and grazing-managed land. The class-area distribution is itself a design signal: the South Island is ~74% class 6–8, so a South Island community’s grazing-protein ceiling is dominated by low-capability land — only ~13% is versatile class 1–3 (which the model routes to veg/cropping first). ⚠ Partial, not a full resolution — see the Research targets and Notes below (North Island legends and the 219-unit extraction remain).

Research targets

Research gaps

• RT_281 (ADVANCED / partial → RD_029; stays OPEN): the South Island Extended Legend (219 units, four South Island MAF regions, Present/Top/Attainable SU/ha DRY+IRR) is now retrieved and filed. Residual: (a) the North Island regional extended legends (the rest of the ~844 units across ~12 legends) are not yet retrieved; (b) the 219-unit South Island table is not yet extracted into the engine’s per-unit/per-region su_ha lookup — a downstream data-processing + per-cell OCR-verification task, not done in this ingest. Keep the RT open, annotated with RD_029.

Notes

Provenance / filing. Downloaded authoritative government publication (MPI document 32359 per the RT), not AI-prepared → no retrieval-provenance block. Single-file PDF → rd_ (national reference dataset, source-files/03_data/). 231 pp scan; Producer “Adobe Acrobat 8.1 Paper Capture Plug-in” (i.e. OCR over a 1983/84 scan). Editor P. van Berkel; NWASCO / Water & Soil Division, Ministry of Works & Development. sha256 5faec79f… recorded for lint L22/L25.

data_quality = verified — scope of that claim. Every figure quoted in key_claims traces to the raw as printed, and the load-bearing anchor (1c1) is double-confirmed (identical in the Example prose p.9 and the detailed table p.11). The verified rating covers these registered claims + the method, NOT the full 219-unit dataset: because the file is an OCR’d scan, any downstream bulk extraction must re-verify each unit’s SU/ha per cell before it reaches an NI calculation (individual OCR digit errors are the risk). The Figure 1 class table is internally coherent (Σha ≈ 15.1 M ha ≈ South Island land area; classes 6/7/8 percentages recompute cleanly). ⚠ The OCR caveat applies to all nine printed percentages, not just the two flagged classes: every % in the Figure 1 table is OCR-read from a 1983 scan and must be treated as approximate (not digit-exact). The discrepancy is largest and demonstrable for Class 1 (printed 0.27% vs ~0.24% recomputed) and Class 5 (printed 0.87% vs ~0.77%), but none of the nine printed percentages is verified exact — flagged, not resolved, as an OCR/rounding artefact.

D25 seam. model_design.md decision D25 states the GIS sampler’s ccav polygon attribute “resolves RT_281 from GIS.” RT_281 as written (and model_design.md §4, the RT_281 gap bullet) instead defines the target as the PDF-legend per-unit extraction — which is this document. Treat RD_029 as the authoritative PDF source RT_281 named; whether the GIS ccav field already delivers equivalent per-polygon carrying capacity is a separate question the extraction step should reconcile.

Cross-references. Method/scope parent: OT_083 (LUC Survey Handbook, Table 23 bands). Hard sheep-beef anchor: OT_082 (7.5 SU/ha hill country). Consumer of su_ha[LUC]: the D02 food module (④f).

Connections

Links to

Sources (2): OT_082 · OT_083

Referenced by

Sources (1): OT_083

EDT domains (1): D02: Smart Food Systems & Agriculture