CR_059: Land-use footprint of ground-mounted solar PV farms in NZ (July 2026 web synthesis)

CR_059: Land-use footprint of ground-mounted solar PV farms in NZ (July 2026 web synthesis)

Land-use footprint of ground-mounted solar PV farms in New Zealand (July 2026 web synthesis)

Advances RT_265 (NZ ground-mount PV land-use factor) and RT_304 (ground-PV density param pv_ground_kwp_per_ha) — neither resolved. Cross-refs agro_pv (the pv_ground_kwp_per_ha=150 premise flag raised in the 2026-07 agri-PV batch).

AI-compiled, medium. The solar figures are primary-verified in this same batch: Bolinger array-density and Lincoln/Massey “1.5 ha/MW” and the Tauhei consent areas were each grep-checked against their retrieved primaries (LIT_120, LIT_121, REG_034). The individual non-Tauhei NZ project rows (Kohirā, Lauriston, Greytown/Far North, Edgecumbe) are news/council-corroborated compiled figures, not retrieved consent primaries — treat them as medium compiled evidence, not model-eligible until each project’s consent is retrieved.

Summary

An AI-compiled synthesis assembling New Zealand’s consented and operational utility-scale solar-farm areas to derive a land-use footprint for ground-mounted PV, for use in the NI land-footprint / ground-PV density calculation. Its core argument is a definitional distinction the model must respect: the international proxy density surfaced elsewhere in the corpus (~870 kWp/ha ≈ 1.15 ha/MWp DC, Bolinger & Bolinger 2022, carried on agro_pv via LIT_113) is an array / direct-impact footprint (panels plus immediate equipment only), whereas NZ resource-consent land areas are total-SITE figures that also include inter-row spacing, access tracks, setbacks, substations, and buffer/wetland/grazing margins. Across five consented NZ projects the total-site footprint runs ~1.3–2.5 ha/MWp DC (median ≈1.5), matching the Lincoln/Massey “1.5 ha per MW” NZ planning rule-of-thumb. The synthesis therefore recommends the NI ground-PV land-density parameter, when it represents a total site, be set to roughly 650–750 kWp/ha (≈1.3–1.5 ha/MWp) rather than the array-only ~870 kWp/ha. (Note: the engine’s current interim value pv_ground_kwp_per_ha=150 sits below both bands — see NI relevance.)

Key claims

- claim: "NZ TOTAL-SITE footprint range — ground-mounted utility solar. Across five consented/operational NZ projects the total consented-or-fenced SITE footprint is ~1.3–2.5 ha/MWp DC (~1.8–3.5 ha/MW AC), cross-project median ≈1.5 ha/MWp DC. Total-site = panel array PLUS inter-row spacing, access tracks, setbacks, substations, and buffer/wetland/riparian/grazing margins. This is the land the community actually alienates, and is the correct basis for a land-footprint (as opposed to array-only) parameter. Only the Tauhei row is a retrieved consent primary in this batch [REG_034]; the Kohirā/Lauriston/Greytown/Edgecumbe rows are news/council-corroborated compiled figures."
  source_location: "cr raw '## Key Finding' + '## Consented and Operational NZ Solar Farms' table; Tauhei area verified in [REG_034]"
- claim: "NZ sector rule-of-thumb (PRIMARY-VERIFIED). 'Given that 1.5 ha (approximately) is required for 1 MW of ground-based solar,' an 8,500 MW addition projected by Concept Consulting for 2027 would need ~13,000 ha of solar-farm land — establishing 1.5 ha/MW as the working NZ planning benchmark. Verbatim, Page, de Koning, Tozer & Ramilan (2024), NZIPIM Journal p.2."
  source_location: "cr raw '## Sector-Wide NZ Rule of Thumb'; verified verbatim p.2 of [LIT_121]"
- claim: "International proxy is ARRAY-ONLY, not total-site (PRIMARY-VERIFIED, with correction). Bolinger & Bolinger (2022) report 2019 US reference power densities of 0.35 MW-DC/acre (≈870 kWp/ha; reciprocal 2.8 acres/MWp in-source — the cr's '2.86 acres' is a rounding) for FIXED-TILT and 0.24 MW-DC/acre (4.2 acres/MWp ≈ ~590 kWp/ha) for SINGLE-AXIS TRACKING, measured from satellite polygons of 736 US utility plants (>5 MW-AC, built 2007–2019). This methodology measures only the array/direct-impact area and explicitly EXCLUDES total leased/owned site: the study's own example (Copper Mountain 3, Nevada) shows the array = 945 of 1,375 acres = 69% of the leased site. Hence ~870 kWp/ha is an array-footprint density (a lower-bound proxy), not a total-site density. NB the retrieved file is the authors' LBNL summary slide-deck (same authors/figures as the IEEE J. Photovoltaics article); ha conversions are the cr's own (source is in acres)."
  source_location: "cr raw '## International Benchmark (Proxy)'; verified in [LIT_120] ('0.35 MWDC/acre (2.8 acres/MWDC) fixed-tilt … 0.24 (4.2) tracking'; Copper Mountain 3 array 945 ac = 69% of site)"
- claim: "Tauhei Solar Farm — consent areas (PRIMARY-VERIFIED). EPA NZ fast-track decision (dated 19 September 2022, COVID-19 Fast-Track Act 2020 — NOT 2025): 262.5163 ha total project land across six parcels vs ~182 ha inside the solar-farm security fence; 186 MW DC / 147 MW AC at consent stage; ~330,000 monocrystalline panels; balance of land retained for wetland (6.9 ha) + riparian (4.8 ha) restoration, biodiversity corridors and sheep grazing. Total-site density ~1.30–1.41 ha/MWp DC (using 202 MWp energised / 186 MWp consented); fenced-array density ~0.98–1.09 ha/MWp DC. The consent explicitly separates total project land from fenced array — the empirical total-site-vs-array split this cr is built on (~0.69 array/total at Tauhei)."
  source_location: "cr raw '## Individual Project Detail' (Tauhei); verified in [REG_034] (decision 19 Sep 2022; 262.5163 ha; ~330,000 panels)"
- claim: "Non-Tauhei NZ project rows (COMPILED, news/council-sourced — NOT retrieved primaries). Kohirā (Kaitāia, Far North): 33.3 MWp DC / 23.7 MW AC on 80–82 ha property (~64 ha array + ~7 ha native reforestation), single-axis tracking → ~2.46 ha/MWp DC (high end; tracking + ecological set-aside). Lauriston (Canterbury): 63 MWp DC / 47 MW AC on 93 ha → ~1.48 ha/MWp DC. Greytown / Far North Solar Farm (South Wairarapa): 175 MWp on 235 ha (Environment Court consent) → ~1.34 ha/MWp DC. Edgecumbe (Bay of Plenty, Genesis, consented not built): 127 MWp on ~207 ha → ~1.63 ha/MWp DC. These bracket the range but each rests on media/council reporting, not a retrieved consent decision — hold as compiled evidence."
  source_location: "cr raw '## Individual Project Detail' + reference list #10–#22 (Lodestone/Infratec/Trina, NZ Herald, PV Magazine, Wikipedia, SWDC, Environment Court, NZX)"
- claim: "Data gaps flagged in the source. (a) Lodestone agrivoltaic projects (Te Herenga o Te Rā / Waiotahe, ~42 MWp DC) are well-documented on capacity but a disaggregated total consented SITE area (vs panel/property area) could not be located in public consent/council documents. (b) MBIE 'Economics of Utility-Scale Solar in Aotearoa NZ' models land-class suitability filters but does not publish a single consolidated ha/MWp planning parameter in the retrievable excerpt. (c) Ong et al. (2013, NREL/TP-6A20-56290) earlier total-plant-area estimate 8–10 acres/MWp (3.2–4.0 ha/MWp) is regarded by Bolinger as outdated given the ~2x power-density rise 2011–2019."
  source_location: "cr raw '## Individual Project Detail' (Te Herenga o Te Rā), '## Sector-Wide NZ Rule of Thumb' (MBIE), '## International Benchmark' (Ong 2013)"

Neobiome Intelligence relevance

The load-bearing contribution is the array-footprint vs total-site distinction for the ground-PV land parameter, and NZ empirical total-site numbers to calibrate it.

• RT_265 (NZ ground-mount PV land-use factor pv_land_ha_per_kwp) — this cr is the most direct evidence for it. RT_265’s placeholder was ~2 ha/MWp; the NZ consented evidence refines the utility-scale total-site factor to ~1.3–2.5 ha/MWp DC (median ~1.5), with tracking + ecological buffers (Kohirā) at the ~2.5 top end and fixed/mixed layouts (Lauriston, Tauhei, Greytown) at ~1.3–1.5. Advances, does not resolve — these are utility-scale projects; a small/community ground array (the RT_265 subject) may pack tighter (less setback/buffer overhead), so the community factor is plausibly at or below the ~1.3–1.5 lower end but is not directly measured here. Stays OPEN.
• RT_304 (ground-array density pv_ground_kwp_per_ha, interim 150) — the reciprocal parameter. Two references now sit above the engine’s interim 150: the array-only ~870 kWp/ha (Bolinger, carried on agro_pv via LIT_113) and this cr’s total-SITE ~650–750 kWp/ha (≈1.3–1.5 ha/MWp). The engine’s pv_ground_kwp_per_ha=150 sits ~4–5x below even the total-site figure — reinforcing (from the land-footprint side) the 🔴 premise flag the agri-PV batch already raised on that constant. Correction the source recommends: when the parameter is meant to represent a total site (land alienated), use ~650–750 kWp/ha; when it represents the array footprint only, ~870 kWp/ha is the proxy. Which definition the engine intends is the open modeling question — not resolved here. RT_304 stays OPEN (its three interim roof factors are also untouched by this source).
• Cross-ref agro_pv — the density section there records the exclusive-use array-only ~870 kWp/ha reference and the pv_ground_kwp_per_ha=150 flag. This cr adds the total-site counterpart (NZ consented ~650–750 kWp/ha) — the same land-footprint premise seen from real NZ resource-consent areas rather than the array-only satellite-polygon method.

See D01.

Underlying sources

• LIT_120 — Bolinger & Bolinger (2022), Land Requirements for Utility-Scale PV: An Empirical Update on Power and Energy Density, IEEE J. Photovoltaics (retrieved as the LBNL summary deck). The array-only ~870 kWp/ha (0.35 MW-DC/acre, 2.8 acres/MWp) fixed-tilt / 0.24 MW-DC/acre tracking benchmark + the Copper Mountain 3 array=69%-of-site example.
• LIT_121 — Page, de Koning, Tozer & Ramilan (2024), The Future of Solar Farming in New Zealand, NZIPIM Journal (Lincoln University / Massey University). The “1.5 ha per MW of ground-based solar” NZ planning rule-of-thumb (8,500 MW → ~13,000 ha), verbatim p.2.
• REG_034 — EPA NZ (2022), Tauhei Solar Farm — fast-track consenting decision (dated 19 Sep 2022). 262.5163 ha total vs ~182 ha fenced; 186 MW DC / 147 MW AC; ~330,000 panels; total-site-vs-array split anchor.
• MBIE, Economics of Utility-Scale Solar in Aotearoa New Zealand — models land-class suitability, no single ha/MWp parameter (already in corpus as OT_056, the Miller 2020 utility-solar analysis).
• Ong et al. (2013), Land-Use Requirements for Solar Power Plants in the US, NREL/TP-6A20-56290 — earlier total-plant-area 8–10 acres/MWp (superseded per Bolinger).
• News/council corroboration for the non-Tauhei project rows: Lodestone/Infratec/Trina (Kohirā); NZ Herald / PV Magazine / Wikipedia (Lauriston); South Wairarapa DC + Environment Court decision (Greytown/Far North); NZX / Genesis (Edgecumbe). Compiled, not retrieved primaries.

Retrieval provenance

• Upstream source: the primaries above — Bolinger & Bolinger 2022 (LBNL deck), Page/de Koning/Tozer/Ramilan 2024 (NZIPIM), EPA NZ 2022 (Tauhei decision), MBIE utility-solar report; plus news/council reporting for the non-Tauhei project areas.
• Prepared by: AI-compiled web synthesis for the NEOBIOME project (deep-research query on the land-use footprint of ground-mounted solar PV farms in NZ; exact tool/prompt unconfirmed) on 2026-07-25.
• Verification: AI-compiled. The solar figures were independently grep-verified against the retrieved primaries in this batch — Bolinger 0.35/0.24 MW-DC/acre + 2.8 acres/MWp + Copper Mountain 69%; Lincoln/Massey “1.5 ha … 1 MW … 8,500 MW … 13,000 ha” (p.2); EPA Tauhei 19 Sep 2022 / 262.5163 ha / ~330,000 panels. Corrections applied: (1) Tauhei consent re-dated 2022, not 2025; (2) Bolinger fixed-tilt reciprocal is 2.8 acres/MWp in-source (the cr’s 2.86 is a rounding); (3) Bolinger file identified as the LBNL summary deck (same figures as the IEEE article). The non-Tauhei project rows (Kohirā, Lauriston, Greytown, Edgecumbe) are news/council-corroborated, not retrieved consent primaries — held as compiled evidence, not model-eligible. No figure hardens a model cell (RT_265/RT_304) until the community-scale definition is settled and the parameter is wired.

Research targets

Research gaps

• RT_265 (NZ ground-mount PV land-use factor pv_land_ha_per_kwp) — advanced by the NZ consented total-site range 1.3–2.5 ha/MWp DC (median ~1.5); stays OPEN (utility-scale evidence; a small/community-scale factor is not directly measured).
• RT_304 (ground-array density pv_ground_kwp_per_ha, interim 150) — advanced by the total-SITE ~650–750 kWp/ha figure (distinct from the array-only ~870); reinforces the existing 🔴 premise flag that the engine’s 150 is well below the literature. Stays OPEN.

Notes

• The cr’s own text asserts the ~870 kWp/ha proxy is “cited in RT_304”. In the corpus that array-only reference is actually carried via RT_108 / LIT_113 on agro_pv; RT_304’s engine parameter is the interim 150 kWp/ha, and RT_265 is the NZ ground-mount ha/MWp land-use factor. This page keeps those straight rather than repeating the cr’s conflation.
• Data-quality caveat: medium. Solar legs primary-verified; the four non-Tauhei project areas are compiled from media/council sources and the community-scale (vs utility-scale) transfer is unresolved.

Connections

Links to

Sources (5): LIT_113 · LIT_120 · LIT_121 · OT_056 · REG_034

Technologies (1): Agro-PV (Dual-Use Solar + Agriculture)

EDT domains (1): D01: Renewable Energy & Storage Systems

Referenced by

Technologies (1): Agro-PV (Dual-Use Solar + Agriculture)

EDT domains (1): D01: Renewable Energy & Storage Systems