Local circumstances

—

— of the Sun covered
by area

— duration

Magnitude—

Sun altitude—

Sun azimuth—

Elevation—

Times shown in local time.

What you'd see

as viewed from your location, zenith up

—

Path of totality

drag to spin · click anywhere to move your location

How it lines up

drag to orbit · scroll to zoom

The Earth, the Moon, their separation and the shadow cone are all to true scale. It opens framed on the Earth, where the cone narrows to a point — that near-miss is the whole reason totality is a few kilometres wide instead of a continent. Zoom out and the Moon comes into frame sixty Earth radii away, which is where it really is. The Sun does not: at 23,455 Earth radii and 109 across, drawing it in place would put everything else inside one pixel, so only its direction is marked. Follows the scrubber above.

Upcoming solar eclipses

Only ones visible from my location

Date	Type	Max duration	Visible from here	Coverage here

How Umbro works

Every number on this site is computed in your browser from first principles. There is no eclipse database behind it, and no API call to fetch.

On iPhone and iPad

The same engine, compiled to Swift and running on the device. It adds the things a web page cannot do: a Live Activity that puts the countdown on your Lock Screen and in the Dynamic Island, Home Screen widgets, notifications for the eclipses you choose, and a camera mode that brackets exposures through the phases of totality. Everything is computed locally, so it works with no signal at all.

Free · iPhone and iPad · requires iOS 17

Why not just use an API?

Because there isn't a good one. NASA/GSFC publishes the authoritative catalogue, but as static pages covering a fixed list of cities. Outreach maps from observatories are images. Xavier Jubier's excellent path files aren't licensed for redistribution. timeanddate.com has the data but no free API. None of them answer the only question that actually matters to you: what happens at my coordinates?

So Umbro carries its own ephemeris and works the geometry out directly. That means it runs offline, costs nothing to serve, and gives the same answer for a mountaintop in Chile as it does for midtown Manhattan.

The method

Solar coordinates come from Meeus's abridged solar theory; lunar coordinates from the truncated ELP‑2000/82 series — sixty periodic terms in longitude and distance, sixty more in latitude. From those two position vectors everything else follows geometrically:

The shadow axis is the line through the centres of the Sun and the Moon.
Gamma is that axis' least distance from the Earth's centre, in Earth radii. It decides whether there's an eclipse at all, and how central it is.
The umbral and penumbral cones follow from similar triangles. Where the umbral cone crosses the Earth's surface, its radius tells you whether you get a total eclipse or an annular one — a negative radius means the cone's apex falls short of the ground, and you see a ring instead.
The central line comes from intersecting the shadow axis with the Earth ellipsoid, stepped through the eclipse thirty seconds at a time.
Your contact times come from solving for the instants when the two apparent discs touch, as seen from your exact spot on a rotating, slightly flattened Earth — not from the geocentre.

This is the Besselian construction, computed on the fly rather than read from published tables of elements.

Percentage covered, and why it isn't magnitude

Two different numbers get called "how much of the Sun is covered", and they disagree badly. Magnitude is the fraction of the Sun's diameter hidden. Obscuration is the fraction of its area — the one that matches what the sky actually looks like, and the one shown in large type on the front page.

At magnitude 0.5 only about 39% of the Sun's area is gone. Umbro computes obscuration properly from the circle–circle lens area, so a "50% eclipse" here means half the light, not half the width.

How accurate is it?

Checked against NASA/GSFC's published catalogue for 2017–2030:

32 of 32 eclipses found, with no false positives
Every eclipse type classified correctly, including the awkward 2023‑04‑20 hybrid that is annular at both ends of its path and total in the middle
Greatest durations within about 3 seconds
Gamma within 0.001 Earth radii

Local circumstances, checked against published figures for individual cities:

Eclipse	Place	Umbro	Published
2024‑04‑08	New York	89.6%	89.6%
2024‑04‑08	Los Angeles	49.2%	49.1%
2024‑04‑08	Dallas	4m07s	3m52s
2017‑08‑21	Chicago	86.7%	87%
2017‑08‑21	Casper, WY	2m28s	2m26s
2026‑08‑12	Reykjavík	1m07s	~1m05s

Contact times land within roughly ±30 seconds. That is comfortably good enough to plan a trip around. If you are timing equipment to the second, check NASA/GSFC too.

Known limitations

Times display in your device's timezone, not the timezone of the place you look up. The label always says which zone is in use, and there's a UTC toggle. Doing this properly needs a timezone-boundary database.
Path edges are approximate by a kilometre or so. The width shown ignores terrain elevation and the Moon's ragged limb profile, both of which shift the exact boundary. Don't plan to stand right on the edge.
IP-based location is rough. If the header badge says "approx", you may be tens of kilometres from where the site thinks you are — which near a path edge is the difference between totality and a deep partial. Search for your town or click the map.

Privacy

All the astronomy runs locally. The only things that leave your browser are map tile requests, place searches you type yourself, and — if you don't grant precise location — a single lookup that turns your IP into an approximate city. No accounts, no analytics, no cookies. Your chosen location is kept in your own browser's local storage and nowhere else.

Looking at the Sun safely

Never look at the Sun without certified eclipse glasses. The only moment it is safe to look with the naked eye is during a total eclipse, between second and third contact, when the Sun's disc is completely covered. During an annular eclipse the photosphere is never fully hidden, so it is never safe — and the same goes for every partial phase of every eclipse. Camera lenses, binoculars and telescopes concentrate the light and will cause injury far faster than the naked eye.

← Back to the tracker

Loading…

—

Today

Local circumstances

What you'd see

Path of totality

How it lines up

Will you actually see it?

Upcoming solar eclipses

How Umbro works

On iPhone and iPad

Why not just use an API?

The method

Percentage covered, and why it isn't magnitude

How accurate is it?

Known limitations

Privacy

Looking at the Sun safely

Privacy

What stays on your device

Your location

Requests that leave your browser

Server logs

Contact

Children, and changes

Terms of use

Eye safety

Accuracy, and what it isn't good enough for

No warranty

Attribution and independence

Acceptable use

Contact