Where these numbers come from
This is an independent guide to the total solar eclipse of Monday 2 August 2027. Every duration, contact time, obscuration, magnitude and Sun altitude on the site is computed from NASA/GSFC Besselian elements by the site’s own engine — not copied from anywhere, and not rounded off into vagueness.
What this site is
On 2 August 2027 the Moon’s shadow crosses Egypt at midday and delivers up to 6 minutes 23 seconds of totality — the longest visible from land anywhere on Earth between 1991 and 2114. This site exists to answer the practical questions that follow from that: where the path actually runs, how long totality lasts in each town, what time it happens on your watch, how high the Sun will be, what the weather usually does in early August, and how to watch it without hurting your eyes.
It is written and maintained by one person as an informational project. There is no organisation behind it, no sponsor, and nothing for sale. Where a fact can be computed, it is computed. Where something is genuinely uncertain — which temple forecourts will be ticketed, what a hotel room will cost in 2027 — the site says so rather than inventing a number.
How the eclipse numbers are produced
Every eclipse figure on this site — the contact times, the duration of totality, the percentage of the Sun covered, the eclipse magnitude, the Sun’s altitude and azimuth — comes from a single calculation, run for the exact coordinates of each place.
The source data: Besselian elements
The input is the set of polynomial Besselian elements published by Fred Espenak, NASA/GSFC, for this eclipse. Besselian elements are the standard compact description of an eclipse: instead of tabulating the Moon’s shadow for every place and moment, they describe the shadow cone’s axis and radii in a coordinate system centred on the Earth, as low-order polynomials in time. Eight of them, plus two cone angles, are enough to reconstruct the whole event anywhere on the planet.
| Eclipse | Total, 2027 August 02 |
|---|---|
| Elements | Fred Espenak, NASA/GSFC |
| Epoch (t0) | 2027 Aug 02, 10:00:00 TDT |
| ΔT (TDT − UT) | 71.7 s |
| Ephemerides | VSOP87 / ELP2000-85 |
| Method | Explanatory Supplement to the Astronomical Almanac, ch. 8 |
| Earth model | WGS-84 ellipsoid, with site elevation |
| Path geometry | NASA/GSFC 120-second path table |
The engine
Those elements are evaluated by js/eclipse-engine.js, roughly 250 lines of
JavaScript written for this site, following the method set out in the
Explanatory Supplement to the Astronomical Almanac. For a given latitude,
longitude and elevation it converts the observer to geocentric coordinates on the WGS-84
ellipsoid, projects them into the fundamental plane, and iterates to the instant of
greatest eclipse and to each of the four contacts. Obscuration is then the overlap area of
two circles; magnitude is the fraction of the Sun’s diameter covered.
One detail worth flagging, because it is where naive implementations go wrong: the Besselian hour angle is tabulated against Terrestrial Dynamical Time, but the Earth’s rotation runs on Universal Time. The engine applies the ΔT correction of 71.7 seconds to the observer’s hour angle before solving. Skip that and every contact time on the site would be about twenty seconds of longitude out.
The path on the map
The shape of the path itself — the central line and the northern and southern limits drawn on the interactive map — comes from NASA’s 120-second path table for this eclipse, on WGS-84, smoothed between table points with a centripetal Catmull-Rom spline. The coloured duration bands (six minutes, five minutes and so on) are not drawn by eye: they are solved by bisection against the same Besselian elements, so the bands and the numbers on the location pages agree with each other by construction.
How accurate is it, and how would you check?
Two claims, both testable against published NASA data:
- The engine reproduces NASA’s published central-line durations to within 0.15 seconds.
- Its contact times agree with NASA’s to the second.
That is the level of agreement you should expect from a correct implementation of the same elements — which is the point. This site is not offering a rival prediction. It is offering NASA’s prediction, evaluated for a few dozen specific Egyptian towns and for any point you care to click on, instead of only for the round-numbered coordinates in a printed table.
The same engine runs live in your browser. On the eclipse map, when you click a point in the Nile valley or drop a pin in the Eastern Desert, your browser is running the identical code with the identical elements and solving the eclipse for that spot in a few milliseconds. Nothing is looked up on a server; there is no lookup table of pre-computed towns behind it.
The honest limits of the calculation
The Moon is treated as a smooth sphere. Real contact times are shifted a second or two by lunar mountains and valleys, which matters most within a kilometre or so of the path edge — if you are chasing the last few seconds of totality near the limit, use a limb-profile prediction, not this one. Times are also computed for each town’s recorded elevation and its listed coordinates; move ten kilometres and the seconds move too. That is exactly what the map is for.
Conventions used throughout
- All Egyptian local times are EEST, UTC+3. 2 August 2027 is a Monday. Locations outside Egypt use their own local time, marked as such.
- First and last contact (C1, C4) are given to the nearest minute; the start, middle and end of totality (C2, maximum, C3) are given to the second, because there the seconds matter.
- Obscuration is the fraction of the Sun’s disc area covered. Magnitude is the fraction of its diameter covered — the two are not the same number, and the difference is why 99% sounds close to totality and is not.
What this site does not do
It is worth being blunt about this, because plenty of eclipse pages exist to sell you something.
- It is not a booking service. You cannot book a room, a tour, a boat or a flight here, and there is nothing to buy.
- There is no commercial relationship with any tour operator, hotel, cruise line or eclipse-glasses vendor — no affiliate links, no paid placements, no sponsored recommendations. When a page says a place is good, that is a judgement about geometry, climate and logistics, not a transaction.
- It does not give travel, visa, security or health advice. Those come from your own government and from the Egyptian authorities, and they change. See Plan your trip for what the site can usefully say, and the disclaimer for what it cannot.
- It does not forecast the weather. The weather pages describe long-run August climatology. A forecast for 2 August 2027 will not exist until a few days beforehand.
- It will not tell you an unfiltered look is fine. See eye safety; use ISO 12312-2 certified filters.
Corrections welcome
If you find an error — a contact time that disagrees with your own calculation, a place described wrongly, a broken link, a claim that has quietly stopped being true — please say so. Astronomical corrections are especially welcome, and if you can show the working, better still. Get in touch; the page will be fixed and, where the change matters, noted.