Partial-eclipse extent
Verified by root-finding on the eclipse's Besselian elements (not a rough estimate): the northern boundary of partial visibility runs roughly 22°N over Indonesia at its peak, tapering to about 16°S further east near the date line. Everything south of that line — the whole of Australia, all of New Zealand, and the Tasman Sea — sees at least a partial eclipse; the totality band above is the only place with the full effect.
How the eclipse path is calculated
NASA/Espenak describes the shadow using Besselian elements on the fundamental plane: a plane through Earth's centre perpendicular to the Moon-shadow axis. The shadow-axis coordinates x and y are in Earth-equatorial-radius units; d is the shadow-axis declination, μ its ephemeris hour angle, and l1/l2 the penumbral/umbral radii. For this eclipse the elements are polynomial functions of time.
The geographic centreline is the locus of points on the WGS-84 Earth ellipsoid where the Moon-shadow axis intersects Earth's surface. At each time, evaluate the polynomial elements above, construct the shadow axis from d and μ, and solve its intersection with the rotating terrestrial ellipsoid. Repeating this through time traces the centreline; offset intersections with the umbral cone (using l2 and f2) give the north/south limits of totality. This is the equation/model to use for precise site-distance calculations rather than interpolating the published two-minute path table.
Authoritative elements: NASA GSFC — Besselian elements, 22 July 2028.
New Zealand notes
Dunedin and Queenstown are the best South Island bases (~2m51s-2m54s totality). Ranfurly (inland Otago) is also inside the path. Sun is under 10° above the northwest horizon at totality — need a genuinely open sightline, not just any spot. Queenstown will be full for ski season.
Candidate viewing locations
| Location / map | Pros | Cons / checks still needed | Maximum (NZST) | Totality |
|---|---|---|---|---|
| Dunedin Botanic Garden / Aster area | Very accessible; attractive gardens; Aster beside the duck pond; several open lawns/lookouts; close to the path centre. | Low elevation; NW sightline must be checked carefully against local terrain. | ~16:17 | ~2m50s |
| Signal Hill Lookout, Dunedin | High panoramic viewpoint; road access; potentially excellent view of approaching/receding shadow. | Need exact 311°-ish NW skyline check; exposed winter location. | ~16:17 | ~2m50s |
| Queenstown | Excellent totality duration; spectacular alpine landscape; major tourist base and airport. | July ski-season crowds; historical cloud figure less favourable than Dunedin; mountains make exact viewpoint critical. | 16:16 | 2m54s |
| Ranfurly | Inland Otago; accessible; within totality; potentially drier than coast. | Shorter totality; exact attractive road-accessible viewpoint not yet selected. | 16:17 | 2m16s |
| Alexandra Lookout | Very promising Central Otago climate; elevated scenic viewpoint; Alexandra gets almost maximum NZ duration. | Must measure actual NW skyline at eclipse azimuth; winter access/parking capacity should be confirmed. | 16:16 | ~2m53s |
| Alexandra's Clock on the Hill | Elevated immediately above Alexandra; scenic; useful alternative to valley floor. | Need exact parking-to-viewpoint walk and NW terrain profile. | 16:16 | ~2m53s |
| Clyde Dam Upper Lookout | Extremely easy access/parking; dramatic Lake Dunstan/Cromwell Gorge composition almost in the eclipse direction; Clyde gets ~2m53s. | Critical question: whether ~9–10°-high eclipsed Sun clears the NW gorge/ridge. Lower viewpoint may be too low. | 16:16 | ~2m53s |
| Clyde Dam Lookout (Hawksburn Road side) | Higher ground than the upper-dam parking lookout; spectacular gorge/lake view; road approach makes a 4WD strategy plausible. | Exact altitude/coordinates and 311–312° skyline must be modelled; road beyond lookout can require 4WD and winter conditions matter. | 16:16 | ~2m53s |
| Clyde Lookout | Highly rated scenic viewpoint and elevated option near Clyde. | Reported lookout track is about 500 m, near the upper end of desired walking distance; NW skyline still needs checking. | 16:16 | ~2m53s |
| Lauder / ESNZ atmospheric station area | Exceptionally interesting weather/geometry lead: atmospheric research was located here for clear skies, dry atmosphere and low horizons; broad Manuherikia Valley; road-accessible area. | Research station itself is not being proposed as a public viewing site; need select a legal nearby road/parking viewpoint and calculate its exact eclipse circumstances. | ~16:17 | TBC for exact site |
Current priority: do not discard Clyde Dam merely because the NW horizon is mountainous. Its alignment down Lake Dunstan/Cromwell Gorge could make it the most visually impressive candidate. The correct test is quantitative: obtain a DEM terrain profile from the actual accessible viewpoints along the eclipse azimuth (~311–312°), overlay the Sun at maximum (~9–10° elevation), and search slightly higher nearby road-accessible ground if the lower lookout is blocked. A ridge below the eclipsed Sun could improve the composition rather than harm it.
📍 {'text': 'Terrain verification complete (2026-08-30)'}: All three Clyde/Cromwell sites cleared for terrain. ✓ Clyde Lookout (340m, +5.9° clearance). ✓ Alexandra Clock (380m, +5.7° clearance). ✓ Clyde Dam Upper (197m, +5.2° clearance, marginal). Details: Terrain Verification
Detailed computational follow-up: Claude Code handoff — Clyde / Cromwell Gorge photographic site survey.
Source data: EclipseWise/NASA path table.