Summary
Perplexity-compiled solar resource assessment for the Tasman/Lower Moutere pilot site (~−41.3°S). Establishes Nelson–Tasman as New Zealand’s premier solar region with annual GHI 1,450–1,550 kWh/m²/yr and specific yield 1,350–1,380 kWh/kWp/yr (bankable P90 ~1,250–1,280). Three independent primary sources (NIWA SolarView, PVGIS ERA5, Global Solar Atlas 2.0) converge on the yield figure. Critical for community PV sizing: the dominant design driver is seasonal swing — January-to-June output ratio 3.7–4.5× by PVGIS solar-angle modelling (or 1.5× by NIWA sunshine-hours, with PVGIS being the more conservative and appropriate basis for off-grid sizing). For a 50-household pilot at the CR_009 design baseline (8,000 kWh/yr/household = 400 MWh/yr aggregate), a Nelson–Tasman-specific yield reduces the community PV array requirement from CR_008/CR_009’s 326 kW (national 14% CF assumption) to approximately 296 kWp (regional ~15.4% CF), with battery storage of 1,500–2,500 kWh for a 1.5–2.5 day winter buffer. Methodologically critical: use PVGIS-derived winter minimum (1.5–2.0 PSH/day in June) rather than NIWA sunshine-hours (2.7 PSH) for off-grid sizing, because PVGIS weights for solar elevation angle at 41°S.
Key claims
- Annual specific yield (Lower Moutere ground-mount): 1,350–1,380 kWh/kWp/yr at optimal tilt, north-facing. Bankable P90 estimate (5–8% downward adjustment): 1,250–1,280 kWh/kWp/yr. Three independent primary sources converge (NIWA SolarView, PVGIS ERA5, Brent 2020 Solar Atlas of NZ in JRSNZ).
- Annual Global Horizontal Irradiance (GHI): 1,450–1,550 kWh/m²/yr (Global Solar Atlas 2.0). Top-tier for NZ. Waimea Plains and Lower Moutere valley — sheltered from westerlies by the Richmond Range — receive similar irradiance to Nelson City itself.
- Annual sunshine hours (Nelson, NIWA 1991–2020 normals): ~2,497 h/yr. Daily PSH range 3.7–4.6 h/day (annual), with summer peak (January) ~4.7 h/day and winter trough (June) 1.5–2.7 h/day depending on methodology.
- PSH methodology — critical for off-grid sizing: NIWA sunshine-hours measures duration of measurable sunshine; PVGIS irradiance models weight by sun angle and atmospheric path length. For battery/generator sizing, use the PVGIS-derived winter minimum (~1.5–2.0 PSH/day in June), not the NIWA figure — PVGIS is more conservative at 41°S because of low winter solar elevation angle.
- Optimal tilt at latitude −41.3°S: 35–36° for maximum annual yield (Solarific empirical optimum); 40–45° as practical compromise for off-grid winter-biased designs (battery-backed community systems). All sources agree on true north (azimuth 0°). East/west costs 10–15% of output.
- Seasonal swing — the dominant battery design driver:
- Summer quarter (Dec–Feb): ~412 kWh/kWp
- Winter quarter (Jun–Aug): ~270 kWh/kWp
- Summer-to-winter ratio: 1.5× by sunshine-hours; 3.7–4.5× by PVGIS solar-angle modelling
- June output is 22–27% of January’s peak (PVGIS basis)
- For comparison: Christchurch (−43.5°) June is 18% of January; Auckland (−36.8°) is 28%
- Anticyclonic winter fog risk (Lower Moutere / Waimea Plains specifically): NIWA Climate of Nelson and Tasman (Macara 2016) notes that winter anticyclones can produce fog in inland Tasman valleys including the Waimea/Moutere area, potentially suppressing generation for 3–5 consecutive days. Battery design must accommodate this — avoid deep discharge in multi-day overcast winter periods.
- Site capacity at 10 ha (theoretical maximum): 6,000–8,000 kWp ground-mount at 60–80 Wp/m² accounting for row spacing and access — far more than 50-household demand requires. Significant headroom for agrivoltaic integration (panels over pasture, horticulture, viticulture) consistent with emerging NZ agriPV research.
- Recommended community sizing (CR_010 framework, for 50-household Tasman pilot):
- Target solar coverage: ~80% of annual demand (with grid/generator backup for winter shortfall)
- Required solar capacity: ~500–700 kWp ground-mount
- Battery storage: 1,500–2,500 kWh usable (1.5–2.5 day buffer)
- Estimated annual generation at 600 kWp × 1,350 kWh/kWp: ~810,000 kWh/yr
- Seasonal backup is not battery-economic: for full mid-winter coverage, a grid connection or backup generator is strongly recommended. Aligns with the Interview II [INT_002] finding (“Grid connection recommended as backup even for self-sufficient communities”) and the OT_001 NZ analysis.
- Three-source data confidence: NIWA SolarView (ground-station + terrain), PVGIS ERA5 (satellite reanalysis), Global Solar Atlas 2.0 (satellite, World Bank/Solargis), plus peer-reviewed Brent (2020) Solar Atlas of NZ in JRSNZ — convergent within ±5% on the headline yield. Strongly recommended next step: run NIWA SolarView at the actual Lower Moutere cadastral address for terrain-adjusted, site-specific irradiance.
Neobiome Intelligence relevance
Closes the supply side of the D01 Tasman pilot sizing model that CR_009 opened on the demand side. CR_009 established 50-household aggregate demand at the conservative baseline = 400 MWh/year. CR_010 establishes regional yield = 1,350–1,380 kWh/kWp/year (≈ 15.4–15.8% capacity factor) — substantially better than the national 14% CF assumption that fed CR_009’s earlier 326 kW estimate.
Updated array sizing for 50-household pilot:
- At national 14% CF (CR_009 estimate): 326 kWp
- At Nelson–Tasman 15.4% CF (CR_010): ≈ 296 kWp for 100% annual energy match
- At CR_010’s recommended 80% coverage: ≈ 500–700 kWp (over-sizes to handle seasonal mismatch and provide curtailable surplus for hot water / EV / agrivoltaic uses)
The OT_018 per-person extrapolation (937 W/person × 140 persons ≈ 131 kW) presumes the high-efficiency passive + heat-pump efficiency stack reducing per-household demand to ~5,500–6,500 kWh/yr. At that demand level and Nelson–Tasman yield, the array would be ≈ 200 kWp. The wiki now holds three array-sizing anchors (131 / 296 / 500–700 kWp) corresponding to three different demand and coverage assumptions.
Battery design driver — the seasonal swing: CR_010 makes clear that the dominant constraint is not daily/weekly variation but the 3.7–4.5× summer-to-winter swing combined with 3–5 day fog events. Battery sizing for 1.5–2.5 day buffer (1,500–2,500 kWh) handles short-term variation; full winter coverage is not battery-economic and requires grid/generator/biomass backup. This converges with the existing D01 evidence stack (Interview II [INT_002] grid-backup principle; OT_001 NZ community battery sizing).
Methodological NI insight: for NI energy yield modelling at southern latitudes (NZ generally, but acutely at 41°S+), use PVGIS solar-angle-weighted PSH, not NIWA sunshine-hours. The 1.5 vs 2.7 PSH June discrepancy propagates directly into battery sizing — a ~45% difference in the binding constraint. Encode this in any NI energy skill that consumes regional resource data.
Site-specific verification step (still pending): CR_010 recommends running NIWA SolarView at the exact Lower Moutere cadastral address to capture terrain-adjusted irradiance accounting for shading from the Richmond Range. This is a high-priority practical RT (RT_099 below) before the project moves from indicative to detailed sizing.
Research targets
Documents to retrieve
Primary references behind headline numbers — CR_010 is a Perplexity synthesis, so figures used in D01/I07 need grounding in primary sources:
-
[RT_099] NIWA SolarView site-specific run for Lower Moutere coordinates — primary NZ irradiance data with terrain adjustment for the actual cadastral address. Free, publicly available at data.niwa.co.nz/solarview. Strongly recommended by CR_010 as the binding next step before detailed sizing. →
url_NNNorrd_NNNcandidate. → D01, I07, NI design -
[RT_100] PVGIS ERA5 simulation at site coordinates (−41.3°S, 173.15°E) with fixed tilt 40° and azimuth 0° (north). Provides P50 / P90 monthly generation estimates suitable for feasibility modelling. WebFetch-retrievable. →
url_NNNorrd_NNNcandidate. → D01, I07 -
[RT_101] Brent A.C. (2020) — “Solar Atlas of New Zealand from satellite imagery” — Journal of the Royal Society of New Zealand, doi:10.1080/03036758.2020.1763409. Peer-reviewed primary for NZ-specific solar resource using commercial satellite data. → RESOLVED → LIT_071 (ingested 2026-07-17). ⚠ Citation scope: LIT_071 tabulates GHI (kWh/m², horizontal), NOT PVOUT specific yield (kWh/kWp) — so it corroborates the accuracy of the satellite dataset behind CR_010’s headline, NOT the 1,350–1,380 kWh/kWp specific-yield figure directly (CR_010’s specific yield rests on NIWA SolarView + PVGIS + Global Solar Atlas PVOUT, so Brent 2020’s role in the ‘three independent primary sources converge’ claim is dataset-accuracy, not a kWh/kWp reading). → D01
-
[RT_102] NIWA Climate of Nelson and Tasman (Macara, 2016) — primary source for the anticyclonic winter fog finding (3–5 day suppression events in Waimea/Moutere area), regional seasonal patterns, and the latitude-vs-irradiance reconciliation for Nelson vs Christchurch/Auckland. PDF at niwa.co.nz. →
lit_NNNcandidate. → D01, I06, nz_specific -
[RT_103] MBIE Economics of Utility-Scale Solar Forecast in Aotearoa New Zealand (Miller, 2020) — primary NZ-government commissioned analysis of utility-scale solar economics; cross-validates Global Solar Atlas 2.0 against NIWA. →
lit_NNNorrd_NNNcandidate. → D01, I01
Connections
Links to
Referenced by