Geolocating Ground Observers from Single Photographs of Total Solar Eclipses Using Lunar Limb1 Topography and Contact Geometry

When the Moon passes directly in front of the Sun, it does not simply slide over the solar disk like a smooth coin covering a lightbulb. The edge of the Moon is not a perfect circle; it is a rugged landscape of mountains and deep valleys. As the Moon moves across the Sun during a total eclipse, sunlight squeezes through the low points between these lunar peaks, creating a string of brilliant, diamond-like sparks known as Baily's beads. For over a century, astronomers have used these fleeting moments to measure the size of the Sun or to time the exact moment of contact, but they have always assumed the observer's location was already known. The question of whether a single photograph of these beads could reveal where the photographer was standing has remained largely unexplored, despite the fact that the pattern of light changes dramatically depending on where you stand on Earth.

A new study demonstrates that a single photograph of a total solar eclipse, paired with the time it was taken, is enough to pinpoint the photographer's location to within a few tens of kilometers. The research, led by computer scientist Sadiya Ahmad at Iowa State University, treats the photograph not just as a picture, but as a precise measurement of the geometry between the Earth, Moon, and Sun. By comparing the specific pattern of light and dark seen in a photo against a computer model of the Moon's bumpy edge, the system can calculate exactly where on the ground that image was captured. This approach turns a casual snapshot into a powerful tool for finding a location without needing any GPS signal, relying instead on the predictable motion of celestial bodies and the rugged topography of the Moon.

The core of this discovery lies in understanding what information is actually present in the image. A common intuition might suggest that the shape of the lunar edge itself—the specific arrangement of peaks and valleys—acts like a fingerprint that changes as you move across the Earth. However, the researcher proved this idea wrong. They calculated that moving 100 kilometers on the ground changes the apparent shape of the Moon's edge by an amount so tiny it is invisible to even the best cameras. In contrast, that same 100-kilometer shift changes the apparent position of the Sun relative to the Moon by a massive amount, shifting the entire pattern of beads and the width of the glowing arc of light. The study confirms that the location signal comes entirely from this shift in alignment, not from the shape of the mountains themselves. The mountains simply determine exactly where along that glowing arc the beads appear, acting as a fixed map that the shifting Sun illuminates.

To test this theory, the researcher built a pipeline that simulates what an eclipse should look like from any point along the path of totality. They used a detailed 3D map of the Moon's surface, derived from satellite data, and combined it with precise calculations of the Moon's orientation and the Sun's position. For every candidate location on the ground, the computer generates a prediction of the bead pattern and the width of the bright arc of light that remains visible just before and after totality. They then compared these predictions against thousands of real photographs. The researcher analyzed 6,973 unselected images from the 2017 total solar eclipse across the United States, a collection of photos taken by the public with consumer cameras that often had inaccurate clocks and no recorded location data.

The results were striking. The system successfully located the photographer in nearly every case, with a median error of 66 kilometers. This means that for half of the photos, the calculated location was within 66 kilometers of the true spot. Even more impressively, not a single one of the nearly 7,000 images was placed more than 1,000 kilometers away from the correct location. The method worked even when the researcher did not know the camera's orientation or the exact time the photo was taken, relying instead on the physical width of the bright arc of light to constrain the possible locations. To verify that this was not a fluke specific to one eclipse, they tested the same system on 190 photos from the 2024 eclipse, which followed a different path and had different lighting conditions. The system again succeeded, finding the correct locations with a median error of 125 kilometers, confirming that the method works across different eclipses and geometries.

One particularly compelling test involved a single, high-quality photograph taken at a known location in Indiana during the 2024 eclipse. The researcher fed this one image into their system, searching the entire continental track of the eclipse from Mexico to Newfoundland without giving the computer any hint of where to look. The system narrowed the location down to a strip along the eclipse path and pinpointed the photographer to within 16.8 kilometers of the true site. This case study highlighted the nature of the solution: the photograph fixes the observer to a specific line across the Earth, with the exact position along that line being the only remaining uncertainty.

The study also explored the limits of this technique. The researcher found that the accuracy is currently limited by the "noise" in the measurements, such as the slight blur of the camera lens or the uncertainty in the exact time the photo was taken, rather than by any flaw in the physics of the method. They estimate that with better timing or multiple photos taken in sequence, the accuracy could improve to within 30 kilometers or less. The work rules out the idea that the shape of the lunar edge alone can be used for location, showing instead that the power lies in the changing angle between the Sun and the Moon.

This research opens up new possibilities for science and history. It suggests that the vast archives of citizen science eclipse photography, often dismissed as mere souvenirs, contain hidden quantitative data that can be unlocked. The method offers a way to verify the location and time of historical eclipse images, providing a tool for forensic verification of past events. It also provides a form of positioning that works without satellites, using only the sunlight and the Moon's shadow. By proving that a single snapshot can reveal a location with such precision, the study transforms the way we view these celestial events, showing that the brief, dazzling moment of an eclipse holds a map to the observer's place on Earth.