The Earth-Shadow Deficit and the Replication Dispute

Between 1949 and 1957, the first Palomar Observatory Sky Survey photographed the northern sky on glass plates.

These plates were the astronomical photographs of their time. Instead of a digital camera, the telescope focused light onto glass coated with a light-sensitive chemical layer. When a plate was developed, each star appeared as a small point. Because the plates survive, researchers can scan them with modern instruments and compare that earlier sky with the sky we see today.

The VASCO research project assembled a catalogue of more than 106,000 transient candidates from those plates. A transient candidate is a point that appears on an old exposure but cannot be found in modern surveys. It may represent a brief event in the sky. It may also be a defect or contamination on the photographic plate. Distinguishing between those possibilities is the central dispute.

An international research team led by astronomer Beatriz Villarroel then asked where those points appeared in relation to Earth's shadow.

Earth does not cast its shadow only across the ground. On the night side of the planet, the shadow continues far into space as a long cone. Objects orbiting high above Earth can still be illuminated by the Sun even while the ground below them is dark. But when an object enters the shadow cone, Earth blocks the sunlight from reaching it.

Many things can produce a point on an astronomical photograph. A distant star produces its own light. It will remain visible whether or not the region of near-Earth space in front of it lies inside Earth's shadow. A defect on the photographic plate should also have no relationship to where that shadow falls.

A nearby object seen only because it reflects sunlight is different. When it enters Earth's shadow, the sunlight is cut off. The object may remain in exactly the same place, but its reflected flash disappears. For every Palomar plate, the researchers used its date, exposure time, and position in the sky to calculate where Earth's shadow would have appeared at different possible distances from the planet.

They could then ask whether the recorded points were distributed randomly, or whether they became rarer where reflected sunlight should become impossible.

That is the pattern the researchers reported. Far fewer transients appeared inside the calculated shadow than the amount of photographed sky would predict. The strongest match occurred at approximately 36,000 kilometers above Earth, the region where geosynchronous communication satellites orbit today.

The team published the result in the peer-reviewed journal Publications of the Astronomical Society of the Pacific in 2025.

The analysis does not measure the distance to any individual flash. It finds that, across eight years of pre-Sputnik observations, the reported population behaved less like distant stars or randomly distributed marks on glass, and more like sunlight reflecting from sources at satellite altitude.