Hubble and Webb found 27 of the faintest trans-Neptunian objects ever directly seen, including bodies only a few kilometres wide—and unexpectedly found fewer tiny objects than some planet-formation models predicted, suggesting the Kuiper Belt was not ground down by collisions in the way astronomers expected.

Twenty-seven faint moving points have taken the directly detected population of trans-Neptunian objects into a size range that was previously governed mostly by extrapolation. The objects were discovered with the James Webb Space Telescope, while coordinated Hubble observations supplied visible-light measurements for part of the sample. Together, the telescopes turned a very small patch of sky into a test of how the Kuiper Belt formed and how thoroughly collisions later reshaped it.

The work is one observing programme reported across two linked papers, not a complete census of the outer Solar System. The finding is worth taking seriously, but it should not be read as the final word.

Webb found 27 objects; Hubble recovered 13

Marielle Eduardo and colleagues report in the size-distribution paper that Webb’s Near-Infrared Camera detected 27 TNOs across only 0.05 square degrees. Their search reached a 40 per cent detection threshold at F150W2 magnitude 28.8, roughly equivalent to magnitude 29.8 in the optical r band. That makes it the deepest Solar System survey reported to date.

The companion colour paper led by Anastasia Morgan recovered 13 of the Webb detections in Hubble images. Combining Hubble’s optical measurements with Webb’s infrared photometry produced colours spanning 0.35 to 3.2 micrometres. For objects that cannot be resolved into discs, those colour measurements provide clues to how their surfaces reflect light and, indirectly, what materials may be present.

So “directly seen” needs a precise meaning here. No photograph reveals a tiny icy landscape. Each object remains a point of reflected sunlight whose motion distinguishes it from the much more distant background.

Finding a moving point beneath the noise

The researchers used shift-and-stack processing. They shifted many exposures according to a large set of plausible orbital motions and combined them. At the correct motion, a genuine TNO reinforces itself in the stack; stars, galaxies and much of the noise do not. A machine-learning classifier then helped reject false positives generated during the search.

The faintest detection reached F150W2 magnitude 29.3. A NASA account published on September 8 says the smallest body was about five kilometres wide. The research paper describes its faintest object as roughly 10 kilometres across when a 15 per cent albedo is assumed. Those figures are a useful reminder that the brightness is measured, while the diameter is inferred. A smaller, shinier object can return as much light as a larger, darker one.

Fewer tiny objects than some models predict

Eduardo’s team found that the number of objects increased relatively slowly towards fainter magnitudes. The nominal sample was fitted by a power-law slope of 0.29, with an uncertainty of plus 0.08 and minus 0.07. The survey therefore did not reveal the sharply rising abundance of tiny bodies produced by some planet-formation models.

Size distributions are historical records. Planetesimals were assembled from dust, pebbles and larger concentrations in the young Sun’s disk. Later impacts can break them into smaller pieces, gradually feeding the low-size end. A population heavily processed by this “collisional grinding” can look different from the population produced at birth.

The result does not mean that Kuiper Belt objects never collided, nor does it identify one model as uniquely correct. It says that this deep pencil-beam survey found fewer very small bodies than some models expected. The simplest picture of a belt steadily ground into an abundant reservoir of fragments is incomplete.

Hot and cold describe dynamical histories

The detections were divided into dynamically cold and dynamically hot groups. Cold classical objects tend to occupy low-inclination, relatively circular orbits and are commonly treated as residents of the belt’s original outer region. Hot objects have more inclined and eccentric orbits; many models place their formation nearer the present positions of Uranus and Neptune before giant-planet migration pushed them outward. Both groups are physically frigid.

Despite those different histories, the two discovery subsets were consistent with the same luminosity-function slope. If larger samples preserve that resemblance, planetesimal formation may have produced a similar spread of sizes under rather different disk conditions. For now, “consistent with” matters: only 27 detections were available, and the split groups are smaller again.

Colour suggests the smallest survivors remember their origins

The Hubble-Webb colours tell a related but separate story. The small cold classical objects occupied the same narrow reflectance sequence seen among larger cold classical TNOs. The dynamically excited objects showed a broader colour range, again resembling their larger relatives.

That continuity is difficult to square with a simple expectation that repeated fragmentation should make the smallest bodies’ surfaces look systematically different. Perhaps collisions were less frequent than expected, or perhaps fragments retain primordial material well enough for the old colour families to survive. The observations do not yet decide between those possibilities.

This surface evidence complements Webb spectroscopy of larger trans-Neptunian objects, which has mapped distinct chemical groupings, and Hubble work suggesting that some Kuiper Belt systems preserve primordial multiple-body structures.

The strongest conclusion is also the narrowest

The survey’s great strength is depth, not breadth. Its 0.05 square degrees sample one tiny line of sight through a broad, structured population. Orbits inferred from short observational arcs can leave classification uncertain, and every conversion from brightness to diameter depends on how reflective the unseen surface is.

Even so, Webb has opened a part of the size distribution that ground-based telescopes could not reach, and Hubble has connected some of those detections to the colour families of much larger bodies. In this field, a point of light is enough to carry history. The first 27 suggest that the Kuiper Belt’s smallest directly detected survivors are not merely anonymous collision debris, but objects whose abundance and surfaces still preserve information about the disk in which they formed.