How we know: methods and sources

We compute every number on this site ourselves from primary sources, and then we check the results against independent published figures. This page explains how.

Eclipse times and length of totality

We calculate the positions of the Sun and Moon as seen from each place, using the Jet Propulsion Laboratory's DE440 ephemeris, which is the same set of planetary tables that NASA missions use. From those positions we find the exact moments when the Moon's disc covers the Sun's, and for the size of the Moon we use the standard value that NASA adopts for its eclipse predictions.

We checked our results against NASA. NASA's eclipse tables, compiled by Fred Espenak, list the length of totality on the centre line every two minutes along the path, with 101 points for 2027 and 82 for 2028. Our figures agree with every one of them to within 0.2 seconds, and NASA's northern and southern limits of the path agree with ours to within 2 kilometres.

Local clock times follow each country's time-zone rules as recorded in the IANA time-zone database when the site was built. Morocco keeps UTC+0 all year from 20 September 2026, and Egypt uses summer time (UTC+3) in August. If a government changes its clock rules before the eclipse, the local times on this site shift by the same whole hour, while the eclipse itself does not move.

Accuracy near the edge of the path

Close to the edge, the Moon's mountains and valleys can add or remove a few seconds of totality. The exact position of the edge also depends on the Earth's rotation in the eclipse year, which nobody can know to the second in advance. Where a town lies within a few kilometres of the edge we say so on its page, so if you plan to watch from near the edge, it is wise to move a few kilometres toward the centre line.

How we work out the clear-sky chance

For every town we looked at the weather at the town's own eclipse hour on every day of the two weeks around the eclipse date, in every year from 1991 to 2025, which gives more than 500 readings per town. The clear-sky chance is the share of those readings with cloud cover of 30% or less. The data come from the ERA5 hourly records of the Copernicus Climate Change Service, on a grid about 25 km across, so a coastal town and the hills behind it may share one number.

These figures describe past weather and are not a forecast. When we compare them with the satellite cloud studies of eclipse meteorologist Jay Anderson (eclipsophile.com), ours show the same pattern but run slightly cloudier, so it is best to treat them as cautious.

How we work out expected totality

Expected totality is the length of totality multiplied by the clear-sky chance, which gives roughly how much totality you can expect to actually see. That makes it easier to compare a long but cloud-prone spot with a shorter, sunnier one.

Sources

Ephemeris: NASA Jet Propulsion Laboratory, DE440.
Reference tables: NASA Goddard Space Flight Center eclipse web site (F. Espenak).
Weather: Hersbach, H. et al. (2023): ERA5 hourly data on single levels from 1940 to present. Copernicus Climate Change Service (C3S) Climate Data Store. Contains modified Copernicus Climate Change Service information 2026. Neither the European Commission nor ECMWF is responsible for any use of this information.
Place names and populations: GeoNames (CC BY 4.0).