Deep Space Tuesdays · your week in deep space

From The Decisional Pass

Monday morning a spacecraft bound for Jupiter will fly into Earth's shadow, spend nearly nine hours running on batteries, and come out the other side 3.5 kilometers per second faster. Nobody is lighting an engine to do it.

The cheapest speed in the Solar System

Juice, ESA's Jupiter Icy Moons Explorer, comes back past Earth on 28 September. Closest approach is 8,640 kilometers up, over the Indian Ocean, at 11:45 UTC. The encounter bends the trajectory by about 20 degrees and hands over that 3.5 km/s for free.

This is the third assist of the trip, after the first ever lunar-Earth flyby in August 2024 and a Venus pass in 2025. One more Earth flyby in January 2029 puts Juice on the line that reaches Jupiter in July 2031, and Ganymede orbit in 2034. Flying straight there needs a much bigger rocket and an unreasonable amount of propellant, twice: once to speed up, once to stop.

Here's the part I keep thinking about. The navigation campaign opened on 17 August with six reserved slots for trajectory corrections, at four weeks, two weeks, one week and three days out among them. They used one.

"Until now, only one manoeuvre at minus four weeks was needed," said Angela Dietz, Juice spacecraft operations manager. "It was of relatively small size, but extremely effective."

They keep the other slots anyway. There's also a final emergency burn held at 08:00 CEST on flyby morning, in case of a navigation error or a collision risk with something in Earth orbit. ESA expects not to need it. You still staff it.

Nine hours on the battery

From 21:24 CEST on 27 September to 06:03 CEST on 28 September, Juice sits in Earth's shadow. It's solar powered, so for those hours it runs on stored charge, and the team has ranked which observations are worth the amps.

The rest is calibration you can't buy. All 10 science instruments run from 23 September to 3 October, pointed at a planet and a moon already known to absurd precision, which is how you learn what your instrument is really telling you before it matters at Europa. Juice also spends several days inside Earth's magnetotail.

Project scientist Claire Vallat calls this one much simpler to operate than the 2024 lunar-Earth flyby, which means they can turn the spacecraft freely and point instruments wherever they want.

The navigation camera gets the assignment I like best: point at the Moon's horizon and test a new optical navigation technique. That camera's real job, years from now, is steering around Ganymede and Callisto, where the round-trip light delay means nobody on Earth is flying anything in real time.

Roman didn't burn it, so it gets 12 more years

Same theme, different mission. NASA said on 14 September that the Nancy Grace Roman Space Telescope now has propellant for at least 22 years of science. It was designed for 10.

Three pieces. The first midcourse correction on 31 August flew with better than 99% accuracy and used 18 kilograms of the 200 budgeted for it. Roman also launched light, 8,056 kilograms against the conservative 9,800 the propellant budget assumed, so the tanks went up full instead of merely sufficient. And because the first burn was clean, the second one and December's L2 insertion should underspend too. Four years, four years, four years.

Two weeks ago I wrote that the Wide Field Instrument was a couple of weeks out. On 15 September it came on: 300 megapixels across 18 infrared detectors, after 10 days drying out at -65 °C, a cooldown to -143 °C to activate, and on toward -183 °C to work. The filter wheel turned in zero gravity for the first time. The first-photon image is a field of badly out-of-focus stars, on purpose, because the array is still stowed. That's the baseline every sharp image gets measured against.

Fire, not ice

A Yale-led paper in Nature Astronomy asks what the outer Solar System built with in its first million years. Two options: chondrules, millimeter beads of rock melted and cooled fast, and matrix, the cold fine dust carrying water ice and organics. It chose fire. The earliest bodies came out 83% to 92% chondrules.

The measurement is the good part. Nothing undifferentiated survives from that epoch, because bodies forming then packed in enough aluminum-26 to melt completely and erase their own texture. So first author Damanveer Grewal went at iron meteorites instead, with two tracers that both follow matrix: sulfur, and the oxidation state of the iron. Independently they agree on 8% to 17% matrix, lower than any chondrite we hold. Two methods, one answer, which is the only reason to believe it.

The comet's methane was a clock

One more on 3I/ATLAS, our third interstellar visitor. Its methane showed up late, after perihelion, which looked odd. A paper accepted in The Astrophysical Journal argues it's a timer. Cosmic rays strip methane from the outer layers over a long interstellar drift, while methane trapped deeper in water ice survives until heating reaches it. Run the thermal model 12,000 times and the runs that match the data want a birthplace below 20 K and a cosmic-ray exposure of 0.87 to 3.1 billion years. The lateness of a gas is an odometer.

References

• Juice to fly past Earth for third gravity assist — ESA, 21 September 2026
• Fuel Savings Double Potential Lifetime for NASA's Roman Mission — NASA, Ashley Balzer, 14 September 2026
• NASA Activates Roman's Primary Instrument, Checks Out Coronagraph — NASA, Ashley Balzer, 15 September 2026
• From the start, the solar system chose fire over ice to build its first bodies — Phys.org / Yale University, 18 September 2026
• Planetesimal compositions governed by aerodynamic sorting from the onset of Solar System formation — Grewal et al., Nature Astronomy, 2026
• Delayed Methane Detection in 3I/ATLAS: GCR-Driven Subsurface Stratification and Interstellar Exposure Age Constraints — Rawat and Acharyya, accepted in The Astrophysical Journal, posted 17 September 2026

Forwarded this? Get The Decisional Pass in your inbox: