Lost Asteroids

Lost does not mean gone. These objects were found, observed for a few nights, and then not seen again. They are still out there on perfectly ordinary orbits. What was lost is our knowledge of where — the observations were too few and too close together to pin the orbit down, so the predicted position has since spread into a region of sky far larger than any telescope can search.

How an Asteroid Gets Lost

An orbit is fitted to the positions observed. The longer the arc between the first observation and the last, the more tightly the fit constrains the result. A few nights gives an orbit good enough to say where the object will be next week, and useless for saying where it will be in thirty years — the small uncertainties compound every time round the Sun.

JPL expresses this as a condition code, usually written U, from 0 for a precisely known orbit to 9 for one that is barely constrained at all. This page lists near-Earth objects at 8 and above. Nearly all of them share the same history: discovered during a close pass, tracked for days while they were bright, then too faint to follow, and by the next apparition nobody knew where to look.

Longest Missing

110 near-Earth objects currently sit at U ≥ 8, oldest last sighting first, showing 60. Filtered to those classed as potentially hazardous. Show all.

NASA/JPL Small-Body Database, fetched 2026-10-03 13:06 UTC. H is absolute magnitude — a brightness measure that stands in for size, where a smaller number means a larger object. Most of these have no measured diameter at all, because that needs observations nobody managed to make.

The One Worth Knowing About

(1979 XB) was observed for four days in December 1979 and has not been seen since. It is roughly 660 m across, which is large enough to matter, and it is on the impact risk list with potential approaches spread across 2056 to 2113. Both facts follow from the same cause: the arc is four days long, so the orbit permits a wide family of futures, and a few of them intersect Earth.

This is the normal shape of the problem. An object appears risky because it is poorly known, not because anyone has calculated that it is coming. Recovering it would almost certainly remove it from the risk list, as recovery usually does — and nobody can recover it, because nobody knows where to point.

They Do Get Found Again

Being on this list is not permanent. Objects come off it in three ways: a wide-field survey happens to catch one and the orbit is matched to the old observations; someone finds the object on archival images taken before it was ever noticed, which extends the arc backwards and is often more valuable than a new sighting; or a predicted close approach is narrow enough that a targeted search succeeds. As surveys have improved, recoveries have become more common — and the list keeps growing anyway, because the same surveys discover new short-arc objects faster than old ones are recovered.

U ≥ 8 is this page's own threshold for calling something lost, not an official designation. There is no formal line at which an object becomes a lost asteroid; the uncertainty simply grows until a search is impractical, and where that lies depends on how much telescope time anyone is willing to spend.