News
DN
NoodleNews
ANALYSIS
8 AUG 2026
Science · Astronomy

Totality lasts longest where almost nobody lives

An estimated 15.2 million residents lie inside the path of the 12 August total solar eclipse. Nearly all of them are in Spain, at the end of the track, where the eclipsed Sun sits low over the horizon and no place gets two full minutes.

15.2M
Residents inside the path of totality, on a population-grid estimate2
138sec
Longest totality anywhere on Earth — 2 min 18 s, over the Atlantic near Iceland1
2°
Height of the eclipsed Sun above the horizon at Palma de Mallorca, the last land in the path1
Where the shadow falls

A path that is mostly ocean and ice

The Moon's dark inner shadow begins at sunrise over Arctic Russia, arcs over the pole, and comes down on north-east Greenland and the west coast of Iceland before crossing the Atlantic to its last landfall on the northern half of Spain and the Balearic Islands.1 The European Space Agency also places the north-eastern corner of Portugal inside the path.5

Almost all of that track is water or ice. Combining Columbia University's gridded population data with its own eclipse geometry, timeanddate estimates that about 15.2 million people live inside the path, against roughly 980 million who will see at least a partial eclipse.2 The land that gets the longest view is the part with fewest people on it: the greatest duration, 2 minutes 18 seconds, falls on open Atlantic near Iceland at 19:46 Spanish time, and the longest totality on land is on the cliffs of Iceland's Westfjords.14

Exhibit 01
Seconds of totality, along the path
Source: Spanish figures from the Instituto Geográfico Nacional municipal eclipse table and its published notes for A Coruña1; Icelandic sites and the Atlantic maximum as published by Space.com from Xavier Jubier's path model4. Duration depends on distance from the centre line, not on longitude, so nearby towns differ sharply. Spanish locations in accent colour.
Key finding. No location in Spain reaches two minutes. The best on Spanish soil is about 1 min 50 s near the Asturian coast, against 2 min 14 s in Iceland's Westfjords.14
The sunset problem

Six minutes to cross a country

Spain sits at the far end of the track, so totality arrives close to sunset and the Sun is very low. The IGN gives the eclipsed Sun at 12 degrees above the horizon at A Coruña, 8 degrees at Burgos and 2 degrees at Palma de Mallorca — low enough that a ridge line, a tree or a building can end the view. Madrid and Barcelona sit just outside the band; Madrid's maximum comes at 7.2 degrees and is partial only.1

The shadow also moves fast. Across the seven Spanish places for which the IGN publishes second-level timings below, totality starts at Vitoria-Gasteiz and finishes at Formentera 6 minutes 11 seconds later. The whole national event, from the first place going dark to the last place coming back, lasts about as long as the walk to a corner shop.

Exhibit 02
Totality's window, seven places in Spain
Source: Instituto Geográfico Nacional municipal eclipse table, times in Central European Summer Time.1 Bar length is the totality window; the axis counts seconds from 20:27:00, the first of these seven starts. Bars are not a geographic ordering — the umbra crosses Spain from north-west to south-east and its edges reach some places before its centre reaches others.
After dark

The Perseids fall on the same night, not the same peak

A solar eclipse can only happen at new moon, so the night of 12–13 August will be free of moonlight — the best condition a meteor shower can have. The International Meteor Organization puts the Perseid maximum on 13 August and expects rates on eclipse night to build to about 40 an hour by 3am local time, which it calls considerably lower than many published predictions. Recent Perseid returns have run below average, with stronger activity expected from 2027.3

What we don't know

The 15.2 million is an estimate, not a count. It comes from a gridded population surface built on data from 2000 to 2020, counts residents rather than people who will be standing in the path on the day, and traces to a single publisher combining that surface with its own path geometry.2 We found no independent recomputation to check it against, and no country-by-country split of it from any primary or institutional source. Our statement that Spain holds the overwhelming majority rests on that published characterisation, not on a calculation of our own.

Whether Portugal is in the path is unsettled in our sources: the European Space Agency includes its north-eastern corner; the IGN's description of the track does not mention it. The IGN's prose gives Burgos 104 seconds while its own table gives 103, and we print the table figure. Exhibit 02 covers only the places for which we hold second-level IGN timings, so the path's entry over Galicia and Asturias is not charted. Cloud cover cannot be forecast at this range, and it, rather than duration, usually decides whether anyone sees anything.

By the numbers
980M
In the zone that sees at least a partial eclipse2
1905
Last total solar eclipse seen from mainland Spain5
2053
Next one visible from Spain after August 20271
40
Perseids an hour the IMO expects at best on 12–13 August3
Caveat. The strongest objection to this piece's own headline is that duration is not the same as a good view. Reykjavík, inside the path, gets 59 seconds — less than Burgos — and Iceland's August cloud record is far worse than inland Spain's. A reader deciding where to stand should weigh cloud odds and a clear western horizon, not seconds alone.
Go deeper

Left out to hold length

Method. Totality durations for Spanish locations are computed from the Instituto Geográfico Nacional's municipal eclipse table by differencing its printed inicio de totalidad and fin de totalidad timestamps; the x-axis of Exhibit 02 counts seconds from 20:27:00 CEST, the earliest totality start among the seven places charted. All differences were computed in Node and asserted against the source timestamps before any figure was printed. The 6 min 11 s national transit is the span from the first of those seven starts to the last of their ends, not the full Galicia-to-Balearics crossing. Icelandic durations and the Atlantic maximum are taken as published and are not recomputed here. Population figures are estimates from a gridded population surface intersected with the eclipse path, reproduced as published; no independent recomputation was possible.
Sources. Numbered, highest tier first; each inline citation points here.
  1. T1Instituto Geográfico Nacional / Observatorio Astronómico Nacional, El eclipse total del 12 de agosto de 2026 — path description, greatest duration and time, Sun altitudes for A Coruña, Burgos, Madrid and Palma, and the municipal timings table. eclipses.ign.es and astronomia.ign.es, retrieved 7 August 2026.
  2. T2timeanddate.com, 10 Things About the August 2026 Total Solar Eclipse — population inside the path and under the partial zone, derived from CIESIN (Columbia University) gridded population data combined with the publisher's own eclipse geometry. timeanddate.com, retrieved 7 August 2026.
  3. T2International Meteor Organization, R. Lunsford, Viewing the Perseid Meteor Shower in 2026, 6 August 2026 — expected hourly rates, timing of maximum, and the note on recent below-average returns. imo.net, retrieved 7 August 2026.
  4. T3Space.com, Where can I see the total solar eclipse on Aug. 12, 2026? — totality durations for Látrabjarg, Snæfellsjökull, Garður and Reykjavík, and the Spanish maximum near the Asturian coast, calculated from Xavier Jubier's path model. space.com, retrieved 7 August 2026. Used for locations the IGN table does not cover; not used for any Spanish figure the IGN publishes.
  5. T2European Space Agency, eclipse briefing material for 12 August 2026 — inclusion of north-eastern Portugal in the path, and the 1905 date for the last totality over mainland Spain. esa.int, retrieved 7 August 2026.
Corrections. None. Material corrections will be logged here with date.
Figures to 7 Aug 2026. Noodle News · v1.0 · eclipse-path-population-totality.