Science1 distinct publisher3 min readPublished
University of Tasmania researchers bounced a NASA signal off a spent upper stage almost 400,000 kilometres out and found it slightly off where models placed it, which is the working argument for radar in cislunar space.
The Scientist · Science desk
Compiled by The ScientistSomething wrong?How this is made
Begin with the residual, because it is the measurement doing the work. The Tasmanian team describes the stage as slightly off its predicted position [6], which is a direction without a magnitude. Without an offset in kilometres, no one outside the collaboration can judge whether the propagated orbit was fine for its purpose or wrong by enough to move the predicted impact point across tens of kilometres of terrain. The published account dates the two passes to July 26 and 29, led by PhD candidate Oliver White [4], while placing the impact only in the previous month [7], so the propagation time that followed the last fix is unstated. That interval is what turns a small residual into a large miss.
The physics explains why this took a network rather than a dish. A round trip to something almost 400,000 kilometres away takes about 2.7 seconds [1], and echo power falls as the fourth power of range, so a target ten times further out returns one ten-thousandth as much signal [4]. NASA's 70-metre antenna at Tidbinbilla transmitted; Hobart, Ceduna and Yarragadee listened, three receivers to one illuminator [5][5]. White called the returning signal incredibly faint [7].
The second observable is what lifts this above a position fix. Liam Filby, applying processing methods from his own doctoral work, read the object's motion out of the same echo and found it rotating about once every seven minutes [8], roughly 8.6 turns an hour [3]. As Filby put it, the echo carries information about how the object is moving, not only where it is [9]. Guifre Molera Calves, who leads space situational awareness at the university, said the group monitors several rocket bodies in the satellite graveyard but had never before studied one about to strike another celestial body [10].
Ground truth is thinner than it looks. The crater is about 20 metres across [2], roughly 1.5 times the stage's 13-metre length [2], and a crater locates an impact without grading the accuracy of a fix taken days earlier. Records made before, during and after the event by observatories in the United Kingdom, Spain and Latvia, and by ESA's station at Cebreros, have been passed to the Tasmanian team for detailed analysis [13].
The claim worth signing is narrow. A spent stage left to the combined pull of solar and lunar gravity drifted to a place its model did not quite predict [11], the Moon offered no atmosphere to absorb the error [12], and a single radar pass caught the discrepancy [6]. Molera Calves's broader point, that human-made objects now travel well beyond Earth orbit and need to be tracked out there [14], is a policy argument this experiment supports without yet quantifying. Australia joined the Inter-Agency Space Debris Coordination Committee in September 2025 [15]; whether that becomes booked transmitter time is a budget question rather than a physics one.
Ranked by verification strength, evidence, and original report placement.
A 13-meter (43-foot) Falcon 9 rocket stage struck the Moon in the month before publication.
University of Tasmania researchers from the School of Natural Sciences were the first in the world to detect the object by radar, helping refine predictions of where it would land.
On July 26 and 29, Ph.D. candidate Oliver White led two deep-space radar observations that formed part of an international effort to observe the event.
NASA's 70-meter antenna at Tidbinbilla, near Canberra, sent a radar signal toward the rocket body almost 400,000 kilometers from Earth, and the faint echo that bounced back was picked up by radio telescopes in Hobart, Ceduna in South Australia and Yarragadee in Western Australia.
The observations showed the rocket was slightly off its predicted position, giving researchers new data to refine estimates of where it would strike the Moon.
Distinct publishers with included, body-backed reporting in this cluster.
phys.org
1 article · September 2, 2026
Follow any of these and your For You feed starts watching them — no settings page required.
product
A five-watt nuclear heater on Blue Ghost turns lunar night into a procurement problem1 distinct publisher
science
Saarland's metallic-glass team buys five days of weightlessness to measure a melt1 distinct publisher
science
Webb splits Hephaistos's two newest Dyson sphere candidates into M dwarfs and background galaxies1 distinct publisher
product
A deorbit contract with a gate: Elytra and the making of disposal-as-a-service1 distinct publisher
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
One narrator, checkable in principle, unchecked in fact
Named people, named instruments, dated passes and a hard number for the tumble rate — this is a specific account, not a vague one. But every element of it reaches us through phys.org's rendering of the University of Tasmania's own telling, and the two figures that would settle the argument are missing: how far off the prediction was, and who measured the crater. The observation dates are precise while the impact is only "last month".
Real hardware, two nights, one target
This was actually done, and doing it required serious coordination: a NASA deep-space dish on transmit, three Australian receivers, and European observatories handing over impact data afterwards. Australia's move into the international debris committee last September gives it institutional footing. What keeps this in the middle of the range is scale — one object, two passes, a first for the team by its own admission, and the European data still awaiting analysis.
Firsts running ahead of figures
The language reaches further than the numbers do. "First in the world to detect by radar" and a warning about debris beyond Earth orbit are doing more work than two nights of observation on a single stage can carry, and the one result that would justify the case for cislunar radar — the model being wrong — never gets a magnitude. The crater width arrives with no instrument attached to it. None of this is inflated invention; it is a modest, genuine result dressed in the vocabulary of a milestone.
The subject is also the narrator
Two PhD candidates get their techniques and their names into international news, a space situational awareness lead gets a demonstration of relevance, and a national agency that joined the debris committee a year earlier gets a domestic partner to point at. Those are all legitimate interests, and they all pull the same way. Phys.org's role in the chain is distribution rather than scrutiny, which leaves the framing — the world first, the warning about lunar debris — essentially where the university placed it.
Plausible in every part, verified in none
Nothing here strains belief — the physics works, the instruments exist, the drift mechanism is ordinary. The uncertainty is structural: one publisher, one institution's account, no released measurements, and an impact whose date this reporting does not supply. If a second observatory or a lunar imaging result confirmed the crater and the strike time, this would move a long way up.