Space · Sunday, 13 September 2026

01 Briefing what happened

Astronomers found 84 objects of a new kind in old Chandra pictures. They were hiding in the light X-ray telescopes see worst.

A University of Alabama team searched Chandra's public archive for dots that vanish in high-energy X-ray images and found 84 objects of a kind never catalogued, in six galaxies.

objects of a new kind, found in six galaxies

all of them in pictures Chandra had already taken and made public [1]

1999

the year Chandra was launched; it is still working

Ars Technica reports that old operating missions are the main target for NASA's science cuts [5]

1.44

times the Sun's mass, the point at which a white dwarf explodes as a Type Ia supernova

nobody has yet caught one of these stars before it explodes [2]

The lead story — what happened

Astronomers found 84 objects of a kind never catalogued, in six galaxies, using pictures already sitting in the public archive of Chandra, NASA's X-ray telescope. [1]
Chandra was launched in 1999 and is still working. [5]
The objects give off very weak X-rays and a great deal of ultraviolet light, so the team calls them hypersoft X-ray sources. [1]
They were found by keeping the dots that appear in Chandra's lowest-energy pictures and vanish in its higher-energy ones. [1]
The team thinks each one is a dead star, a black hole, neutron star or white dwarf, pulling gas off a companion star. [1]
Pairs like that usually glow in high-energy X-rays because the stolen gas gets very hot. [2] These do not, and the team does not yet know what kind of pair they are. [1]
Nobody saw them before because low-energy X-rays are hard for telescopes to catch, and the hydrogen and helium gas between stars soaks up ultraviolet light. [1]
None has been found in the Milky Way yet. The team's explanation is that from Earth we look through our own galaxy's gas. [2]
Two of the six galaxies are spirals, Andromeda and the Pinwheel; four are ellipticals. The objects sit both where stars are forming and among old stars. [1]
They may explain what strips electrons from the gas between stars, a job hot massive stars are not numerous enough to do on their own. [2]
They may also be the long-sought stars that go on to explode as Type Ia supernovae, the explosions astronomers use to measure how fast the universe is expanding. [1][2]
The paper was published in Nature Astronomy on Wednesday, 9 September. [1][2]

Who is involved

Mustafa Muhibullah

an astronomer at the University of Alabama; he led the study that found the 84 objects
Chandra

NASA's X-ray telescope, in orbit since 1999; its archive of old pictures held the objects
Jimmy Irwin

an astronomer at the University of Alabama and co-author; he wants to catch a Type Ia supernova before it explodes
Rosanne Di Stefano

an astronomer at the Center for Astrophysics, Harvard & Smithsonian, and co-author; she says the search removed a blind spot

How it unfolded

1999 Chandra, NASA's X-ray telescope, is launched [5]
Since then Chandra photographs Andromeda, the Pinwheel and four elliptical galaxies; the pictures go into a public archive [1]
Wed 9 Sep the team publishes 84 hypersoft X-ray sources in Nature Astronomy [2]
Next find one nearby, or catch one before it explodes [2]

Where this points

Watch whether anyone finds one of these objects inside the Milky Way, where our galaxy's own gas has hidden them so far. A nearby one could be studied in detail, not as a dot in another galaxy.

What is pushing on the whole day

The bar and the word are our reading of how hard each one is pushing today. The arrow is where it is heading. The evidence is in the stories below.

Looking in a band nobody had used Building↑

The 84 objects turned up only in Chandra's lowest-energy pictures, the band X-ray telescopes see worst. [1] Webb measured Chariklo's rings in infrared light that Earth's atmosphere blocks from telescopes on the ground. [3] NASA's Moon Base manager says the first phase exists to measure the south pole's shadows, dust and micrometeorites, ground Apollo never visited. [4]

Long gaps between looks Steady→

Chariklo's rings were found in 2013, and the next reading needs another star to pass behind them. [3] Astronomers have looked for years for the stars that become Type Ia supernovae without finding one. [1] NASA does not expect to start permanent hardware on the Moon until 2028 or 2029. [4]

Data anyone can use Building↑

The Chandra images the team searched are openly available to the public. [1] Amateur astronomers' decade of occultation records is what made the change in Chariklo's rings recognisable. [3] NASA's Moon Base page invites companies, universities and innovators around the world to help build it. [7]

The rest of the day

2 more stories on this beat.

Each with its own sources. None of these is a link to the story above.

02
Webb sees an asteroid's rings change

Chariklo is a body 250 km across, orbiting between Uranus and Neptune. [3] In 2013 it became the first small body found to have rings. [3] A team led by Pablo Santos-Sanz of Spain's Institute of Astrophysics of Andalucia used the James Webb Space Telescope to watch a star pass behind the rings, and published the result in Science Advances. [3] It was the first such occultation ever predicted and observed from Webb. [3] Compared with earlier ground-based records, the inner ring now blocks far more starlight and the outer ring has thinned until it is hard to detect. [3]

Why it matters — Rings were thought to belong to giant planets, and nobody expected a 250 km body's rings to change in a decade. The team says the next step is another occultation, and amateur astronomers along its ground track may be part of it. [3]
03
NASA's Moon Base chief sets out three phases

NASA published an interview with Carlos Garcia-Galan, who runs its Moon Base programme, recorded on 14 August. [4] NASA announced the plan for an outpost near the Moon's south pole in March 2026. [7] Phase one is for learning: several companies landing reliably, and measuring the pole's shadows, dust, radiation and micrometeorites, before permanent hardware starts in 2028 or 2029. [4] He hopes for one robotic landing this year and several next year. [4] The Artemis IV crew in 2028 would be the first to meet Moon Base hardware, driving vehicles sent ahead of them. [4]

Why it matters — Habitats must sit at least a mile from where a lander touches down, because its engines blast the ground, so every crew's stay begins with a drive. Humans have spent about 80 hours in total working on the Moon's surface, and this plan would multiply that. [4]

02 Lesson why it matters

What a telescope cannot see looks like empty sky

An instrument catches only some kinds of light. Anything that shines mostly in the other kinds shows up as nothing, until someone looks in a new band.

The twist

Nothing was hiding these 84 objects except the choice of which light to look in. They appeared the moment the team compared Chandra's low-energy pictures with its high-energy ones.

How it works

A telescope is built to catch one range of light
Whatever shines mostly outside that range leaves a faint mark, or none
The catalogue fills up with what the telescope sees well
In that catalogue, empty sky and an unseen object look the same
A new band, or a comparison between bands, turns the blanks into objects

The same force, elsewhere today

Where this chain is also running, in today's other stories.

Webb sees an asteroid's rings change

The earlier readings were made from the ground in visible light and the new one from Webb in infrared. The first thing the team must rule out is that the band changed, not the rings.
NASA's Moon Base chief sets out three phases

Apollo never landed at the south pole, so its shadows, dust and micrometeorites are blanks in the record. Phase one exists to fill them in before anyone builds there.

Where you've seen this

Medical scans

an X-ray shows bone and misses a soft-tissue tumour, so the tumour is absent from the picture until an MRI is done

Fishing surveys

a net with a set mesh counts only fish bigger than the mesh, so the small ones are missing from the count, not from the sea

Crime maps

a police map shows what was reported, so a crime nobody reports is missing from the map, not from the town

The catch

A new band shows what the old one missed and leaves out other things in turn. Webb's infrared cannot see what Chandra's X-rays can, so no single instrument ever gives the full list.

And the whole of it

Everyone reads the world through some instrument: a telescope, a survey, a news feed, their own two eyes. Each one leaves blanks that look like empty space. The people in these stories know where a few of their blanks are. Most of us, the reader included, never find out what ours have left out.

03 Truth what's really going on

What is really going on

Chandra's public archive held these 84 objects for years. X-ray telescopes are built to catch high-energy light, and these objects give off almost none of it, so they sat in the pictures as faint dots nobody had sorted out. Mustafa Muhibullah's team at the University of Alabama found them by keeping only the dots that show up in Chandra's lowest-energy pictures and vanish in the higher-energy ones. [1]

Why it works on us — The words 'mysterious' and 'unlike any they have seen before' make a signal nobody has explained feel like something that has just arrived in the sky, when the objects were in the pictures all along and what is new is the sorting. [1][2]

Who gains

Mustafa Muhibullah's team at the University of Alabama — A new class of object, named by them, out of pictures already in a public archive; no new telescope time was needed. [1]
Chandra, NASA's X-ray telescope, launched in 1999 — A 27-year-old telescope has just produced a new class of object from its archive. Ars Technica reports that operating missions well past their design lives are the main target for NASA's science cuts. [5][2]
Amateur astronomers who record occultations — Santos-Sanz said the decade of ground-based records, some from back gardens, is what let the team recognise that Webb's reading was unusual. [3]
Astrobotic, Intuitive Machines, Firefly and Blue Origin — Phase one of the Moon Base is described as more than twenty landings bought from several vendors, and NASA's events list names their landers under the Moon Base and CLPS labels. [4][6]
Webb's schedulers — The first predicted occultation observed from Webb worked despite a last-minute course correction, and the method was reused on Quaoar. [3]

Who pays

Astronomers who measure the universe's expansion with Type Ia supernovae — They still do not know which stars produce their measuring stick. Jimmy Irwin said they study the explosions only after they have happened. [2]
Scientists waiting on Chariklo's rings — The next reading needs another star to pass behind the rings as seen from Webb, and the team cannot schedule that. [3]
The first Moon Base crews — Garcia-Galan says habitats must sit at least a mile from the landing site because landing engines blast the ground, so every crew begins with a drive across the pole in a vehicle sent ahead. [4]
Anyone hoping for a Moon Base on a fixed date — NASA's programme manager puts the start of permanent hardware in 2028 or 2029, after a learning phase whose landings have not begun. [4]

What nobody knows yet

Open questions from across today’s stories — ours included.

01

What the 84 objects actually are.

The team says they most likely involve a black hole, neutron star or white dwarf pulling gas off a companion star, but has not identified a single one. [1]
02

Whether any exist in the Milky Way.

None has been found in our own galaxy. The team's explanation is that we look through the Milky Way's own gas, which soaks up ultraviolet light, so a nearby one would be hidden by the same effect that hid all 84. [2]
03

Whether these objects are what strips electrons from the gas between stars.

Hot massive stars are not numerous enough to explain all of it, and the team offers the new objects as a possible answer, not a measured one. [2]
04

Whether any of them will explode as a Type Ia supernova.

Astronomers have searched for the stars that become these explosions for years without finding one. The team says these objects could be candidates and no more. [1][2]
05

Whether Chariklo's rings really changed.

The Webb team measured a clear difference from earlier ground-based records. But the rings could be lumpy, so a different slice was measured, or they could look different in infrared than in visible light. Pablo Santos-Sanz says going from the measurement to a cause is much harder. [3]
06

What is moving material between Chariklo's two rings.

Mass passing from the outer ring to the inner one, or an unseen moon shepherding the inner ring, are the two ideas. Only a future occultation can test either, and those happen when the geometry allows, not when astronomers choose. [3]
07

When the Moon Base's first robotic landing flies.

Carlos Garcia-Galan said in August that he hoped for one landing this year and several next year. NASA's own events list names the landers under Moon Base and CLPS labels. [4][6]
08

How long the shadows at the lunar south pole actually last.

Garcia-Galan said it may be several days but not fourteen. Phase one is meant to measure it, because Apollo landed elsewhere. [4]

04 Hope carry this

Eighty-four objects of a kind nobody had catalogued were found this week in pictures that had been sitting in Chandra's public archive, open to anyone.

Also true today

Webb's first predicted stellar occultation worked, and the same set-up has since been used to study the rings of Quaoar, a dwarf planet in the Kuiper belt.
The change in Chariklo's rings could only be recognised because amateur astronomers had been recording occultations of it for the past decade.