Eclipse 2028 practical guide
Dunedin's 1163 Total Solar Eclipse
Otago's previous total eclipse predates any written record from Aotearoa New Zealand. What survives is a calculated date from NASA's catalogue and a modern astronomer's public account of its significance - not an eyewitness page to transcribe.
In brief
NASA's solar-eclipse catalogue lists a total solar eclipse on 27 December 1163 (Saros 118), with its point of greatest eclipse far south in the Southern Ocean. Otago Museum director Dr Ian Griffin has publicly stated this was the last total eclipse visible from Dunedin, making 22 July 2028 the first there in more than 850 years.
What the catalogue establishes
NASA's five-millennium solar-eclipse catalogue lists catalog number 07523: a total solar eclipse on 27 December 1163, Saros series 118, with its point of greatest eclipse near 67 degrees south, 91 degrees east - deep in the Southern Ocean, far from Otago. A catalogue entry for greatest eclipse is a single global reference point, not a statement about any specific city; the umbral path at that date swept across a wide range of latitudes, and whether it crossed a given Aotearoa New Zealand coordinate is a separate calculation from the greatest-eclipse point alone.
Why Dunedin specifically
The connection between that catalogue date and Dunedin comes from a named, current source rather than this site's own path calculation: Dr Ian Griffin, astronomer and director of Otago Museum, has told reporters that 27 December 1163 was the last total solar eclipse visible from Dunedin, and that 22 July 2028 will be the first there since - and the first visible from anywhere in Aotearoa New Zealand since 1965. That reporting is the basis for the "first since 1163" framing used elsewhere on this site for Dunedin.
A necessary caveat on Queenstown
Some regional coverage extends the "since 1163" claim to the whole Otago region, including Queenstown, since both cities sit close to the 2028 centre line roughly 100km apart. That claim traces to a named astronomer quoted on the record for Dunedin specifically - nobody has published the equivalent check for Queenstown's exact coordinates, so we're not stating a date for Queenstown here.
What 1163 and 2028 don't share
No written or oral record of the 1163 eclipse as seen from Otago survives to transcribe - unlike Sydney's 1857 event, there is no newspaper archive to search. The 2028 eclipse will be the opposite: forecast years in advance, observed from named coordinates, and recorded by more instruments and photographers than any eclipse in Aotearoa New Zealand's history. Griffin has also pointed out a specific advantage of the 2028 geometry for Otago: the Sun sits low in the sky at maximum, which favours dramatic landscape photography compared with a higher-altitude eclipse elsewhere.
The 1163 date itself is real and catalogued. What any particular person in Otago actually saw that day, though, is simply lost to history - no amount of research recovers a record that was never written down. A travel article can paper over that gap with invented atmosphere; this page treats the absence of a record as itself worth stating plainly.
Why 865 years passed between the two eclipses
A total solar eclipse recurs at any single location only rarely, because the Moon's shadow traces a narrow path across a small fraction of Earth's surface each time, and successive eclipses in the same Saros series shift roughly 1,200 kilometres west with each 18-year, 11-day cycle. It typically takes many centuries of accumulated Saros cycles, each shifting the path a little differently in latitude as well as longitude, before a path happens to cross the same city again. Dunedin's 865-year gap and Sydney's shorter 171-year gap both illustrate the same geometry - the interval between two total eclipses at one place is essentially a matter of chance intersection, not a fixed rhythm.
This is also why "first total eclipse in centuries" claims recur so often around the world without being exaggerated marketing: most inhabited places genuinely do wait centuries between events, and a NASA catalogue entry for one location says nothing about the wait time for a neighbouring one - which is exactly why the Queenstown caveat above stands on its own rather than borrowing Dunedin's figure.
Saros 118, the series containing the 1163 eclipse, is itself unrelated to Saros 146, the series that produces the 2028 eclipse - two independent numbering sequences that happen to place total eclipses over the same city centuries apart by coincidence of geometry, not because one series descends from the other.
Frequently asked questions
Does the 1163 date mean Dunedin has direct historical records of that eclipse?
No. It is a calculated astronomical date from NASA's catalogue; Aotearoa New Zealand has no written record from 1163, so there is no eyewitness account to cite.
Does the 'first since 1163' claim also apply to Queenstown?
Not on this page's evidence. That framing comes from a named astronomer's public statement about Dunedin specifically; nobody has published the equivalent check for Queenstown's exact coordinates, so this page states a date for Dunedin only.
This page follows the site's correction policy: factual errors are fixed and the update date above changes so the fix is visible, rather than silently edited.
Continue researching
Compare these circumstances with the Dunedin 2028 eclipse guide and Queenstown 2028 eclipse guide, read how eclipse geometry is calculated, see the author's own eclipse-observing background, or compare with Sydney's 1857 total eclipse and the 1922 Wallal eclipse expedition.