CR_013: NZ small-wind capacity factor by region + Lower Moutere pilot (compiled synthesis + verification)

CR_013: NZ small-wind capacity factor by region + Lower Moutere pilot (compiled synthesis + verification)

NZ small-wind capacity factor by region + Lower Moutere pilot (compiled synthesis + verification)

Compiled research — derived estimate, not measured; upstream-referenced

The per-region capacity factors are derived (NIWA station wind speeds → power-law shear → power-curve mapping), not measured — no published per-region small-turbine CF dataset exists for NZ. Numbers are AI-extracted from upstream sources. A separate Verification & enrichment layer (2026-06-05) cross-checks the central conclusion against the BRANZ figure and international field trials; it is annotation, not the source’s own claim. Use wind_cf values tagged assumed.

Summary

A compiled-research synthesis estimating annual capacity factors (CF) for small wind turbines (1–100 kW, ~10–30 m hub height) across all 16 NZ regions, with a specific estimate for the Lower Moutere / Nelson-Tasman pilot site. It anchors on published NIWA station mean wind speeds, scales them to ~20 m hub height via a power-law shear profile, and maps to CF using a Rayleigh/power-curve approximation calibrated to the 2013 Nelson City Renewable Energy Study. The headline finding is that NZ small-wind CF is low (≈0.05–0.10 in sheltered inland regions, 0.22–0.30 only at exposed coast/ridge sites) — roughly a third to a half of NZ utility wind’s ~40% — and that the pilot site sits near the bottom of the national range. Directly relevant to D01 and to the NI energy calculation’s wind_cf regional parameter.

Key claims

- claim: "NZ utility-scale onshore wind averages ~40% annual capacity factor (up to ~50% at the most exposed Tararua/Wellington sites); the Electricity Authority uses 40% as the planning assumption for new onshore wind."
  source_location: "§1.1 NZ Utility Wind Farm Benchmarks"
- claim: "NZ small wind turbines typically deliver about 25% to 30% of their rated output (BRANZ/Level.org.nz); economic viability generally needs ~6–8 m/s mean wind (EECA/Level) — this is the optimistic good-site figure."
  source_location: "§1.2 C / verification (level.org.nz confirmed)"
- claim: "Derived per-region small-turbine CF at 20 m hub: sheltered inland lowest — Waikato 0.05–0.09, Tasman/Nelson 0.07–0.10, Central Otago basin 0.06–0.10; exposed highest — Wellington 0.22–0.30, Taranaki 0.18–0.26, Auckland 0.17–0.22, Southland 0.15–0.21."
  source_location: "§3 Per-Region Table / §5 Summary Table"
- claim: "Lower Moutere (Nelson-Tasman) small-turbine CF best estimate ≈ 0.10–0.16 for a valley/sheltered site and 0.16–0.22 for an exposed ridge, with ±30–40% uncertainty pending a site measurement campaign; the basin representative figure is 0.07–0.10."
  source_location: "§4 Lower Moutere Pilot Site / §3 rows 10–11"
- claim: "Nelson basin wind power density is ~2 W/m² vs >9 W/m² at Wellington (>4×); the 2013 Nelson City RE Study concluded small-scale wind does not stack up economically (LCOE ~$250–500/MWh) except in off-grid niche scenarios."
  source_location: "§1.2 D / §4 (Nelson City RE Study 2013)"
- claim: "Nelson Airport sits below a small turbine's cut-in speed 42.9% of the time; Alexandra (Central Otago) 54.4% — the highest 'calm fraction' in the EA 89-station dataset."
  source_location: "§3 narrative (Nelson #11, Otago #15)"
- claim: "Method: NIWA 10 m station mean wind speeds scaled to 20 m with a power-law shear exponent α=0.20 (×1.149), then mapped to CF via a Rayleigh/power-curve approximation calibrated at V20≈5.5 m/s → CF≈0.20. The CF values are estimates; no published per-region small-turbine CF database exists for NZ."
  source_location: "§2 Derivation Methodology"

Neobiome Intelligence relevance

Fills the regional wind_cf parameter (②), tagged assumed. This source supplies the 16-region small-wind CF layer the energy model needs to assess wind feasibility anywhere in NZ, plus the pilot value. Because the CF is derived rather than measured, the workbook cell should read assumed, not sourced. For the Tasman / Lower Moutere pilot: P50 ≈ 0.10, P90 ≈ 0.07 (low end of each region’s band).

Expected calculation consequence (a feature, not a gap). With wind_cf ≈ 0.10 at the pilot, small wind will correctly come out marginal-to-uneconomic in the portfolio sweep — it only becomes a serious contributor at exposed sites (Wellington, Taranaki, exposed Southland/Marlborough coast). This is the model honestly declining a popular-but-unsuited technology for a sheltered inland site, consistent with the Nelson RE Study’s own conclusion.

Verification corroboration (annotation, not the source’s claim). The central low-CF finding is independently confirmed: BRANZ/Level’s “25–30% of rated” is verified verbatim but is the good-site ceiling; international year-long field trials run lower — UK Energy Saving Trust found free-standing rural turbines averaged ~19% and all building-mounted units <8%, the Warwick rooftop trial a mean 4.2%, and a 2025 study of 173 sites found only ~7.5% exceeded a 10% CF. The report’s 0.07–0.22 NZ range sits inside this measured international band; if anything the low end is optimistic for sheltered inland installs (often 0.02–0.08).

Vertical-axis turbines do not change this. VAWTs are less aerodynamically efficient than HAWTs (Cp ~0.15–0.30 vs ~0.35–0.45); their advantages are mechanical, not yield. Low mean wind caps output regardless of axis, and building/roof-mounting is a recognised failure mode. A VAWT technology will not be added to the model on benchmark evidence alone — a supplier datasheet is required first (RT_202).

Caveat for calculation use. As a cr_ synthesis the numbers are AI-extracted from upstream documents; the load-bearing anchors (NIWA station wind speeds, the Nelson RE Study) should be retrieved directly before the derived CF is treated as anything firmer than assumed (see Research targets).

Research targets

Documents to retrieve

• NIWA national wind-resource dataset (station mean annual wind speeds + regional wind maps; Global Wind Atlas NZ as a free gridded cross-check). The primary speed layer behind every derived CF here; retrieving it would let wind_cf be re-derived from a sourced wind-speed layer rather than AI extraction. Cross-references the open RT_140 (national wind/hydro resource layer) — not minted as a new target to avoid duplication.
• Nelson City Council Renewable Energy Study (2013) — primary for the ~2 W/m² Nelson power-density and “doesn’t stack up except off-grid” conclusion. Not yet retrieved. Cross-references RT_140.

Research gaps

• No published measured small-turbine CF dataset exists for NZ — the per-region CF is necessarily derived. This gap is not closable by retrieval (the data does not exist short of field measurement campaigns); flagged for honesty, not added to the central backlog.
• National micro-hydro resource/availability layer remains open (the wind half of RT_140 is delivered here, derived; the hydro half is not).

Connections

Referenced by

Sources (6): CR_029 · DS_001 · DS_002 · DS_004 · OT_030 · OT_075

Decisions (1): D_002: National scope — all of NZ, no single p…

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

Technologies (1): community-scale)