Moon & Mars Mondays · your week on the Moon and Mars

From The Decisional Pass

Last Monday I told you a spent rocket stage was going to hit the Moon on Wednesday, and that the interesting part would be watching people confirm it.

They confirmed it. A South Korean spacecraft got there first.

Eight passes over a fresh hole

The Falcon 9 upper stage — the one that carried Blue Ghost Mission 1 and Resilience toward the Moon in January 2025 and then drifted for eighteen months — hit near Einstein crater at 06:34 UTC on August 5, which was 15:34 in Seoul.

The impact flash was never really available to us. The stage came down on sunlit ground, and a flash against a lit lunar surface is a flash against a floodlight. Ground telescopes tried anyway. That was always the long shot.

What worked was orbital, and it worked because somebody planned it properly. Danuri — Korea's first lunar orbiter, run by KARI — started imaging about thirty minutes before the collision, and then, thanks to an orbit adjustment made specifically for this, passed back over the predicted impact point repeatedly. Eight imaging sessions in total, using its Lunar Terrain Imager and its wide-angle polarimetric camera.

The result is exactly what you want and almost never get: a before, and a series of afters, from the same instrument, on the same day.

What's in them is a patch of ground that got noticeably darker after the impact, with a bright spot at the center and streaks of ejecta thrown outward — the signature of fresh material excavated and splashed across older, weathered regolith. Preliminary sizing puts the new crater at up to about 27 meters across.

Hold that number next to the prediction. NASA's August 4 advisory expected roughly 60 feet wide and 12 feet deep — about 18 meters. The first measurement came in bigger than the forecast.

I don't want to overread that. "Up to 27 meters" is a preliminary number off a wide-angle instrument, the dark ejecta halo is much larger than the crater itself and is easy to conflate with it, and nobody has had time to do this carefully. But this is precisely why you do the experiment. We had a known mass, a known velocity, a known impact angle, and a documented prediction written down in advance. Now we have an observation that doesn't quite match. That gap is the entire scientific value of the event — it's a free calibration point for how a big, hollow, tumbling metal tube deposits its energy into regolith, which is not the same as how a dense rock does.

Also being processed: data from ShadowCam, the NASA instrument riding on Danuri that normally hunts water ice in permanently shadowed polar craters, from the close pre-impact pass.

LRO is the closer. Its narrow-angle camera resolves features down to about 50 centimeters, and it imaged the target area beforehand to establish a baseline. It's expected to photograph the site in the coming weeks, and KARI and NASA intend to analyze the two data sets together. Korean scientists have already been comparing against an LRO frame from October 2015 — an eleven-year-old picture of an unremarkable patch of the Moon, suddenly the most valuable "before" image in the solar system.

Somebody took that picture in 2015 for no reason at all except that mapping the Moon is what LRO does. That's the whole argument for baseline surveys, right there.

A bad morning at Wenchang

This morning, at about 12:00 UTC, a Long March 7A lifted off from Wenchang carrying the Zhongxing-4B communications satellite, and broke apart roughly 85 seconds into flight. The payload was lost. As of Andrew Jones's report for SpaceNews, Chinese authorities had not yet issued an official account, and it isn't publicly clear whether this was a vehicle failure or the flight termination system doing its job in response to one.

It's the fourth Chinese launch failure of 2026, against 56 orbital attempts.

Here's why it lands on a Moon newsletter. Chang'e-7 — China's lunar south pole mission, with an orbiter, lander, rover, relay satellite and a hopping probe built to work steep slopes — is scheduled for August 24, two weeks from today. It flies on a Long March 5, not a 7A.

But the two vehicles share the YF-100 kerolox engine family, and they share Wenchang. Whether Chang'e-7 holds its date now depends almost entirely on what the failure investigation finds and how narrowly it can be scoped. If the cause is specific to the 7A's configuration or to one bad unit, you can fence it. If it points at the engine family, or at manufacturing, or at pad operations, the fence gets much wider and a lot of the Chinese manifest moves with it.

This is the least glamorous and most consequential kind of aerospace decision: how much of a fleet does one anomaly contaminate? Scope it too narrowly and you fly into the same failure. Too broadly and you ground a national program over a part that was fine.

Chang'e-7 is aiming for the sunlit rim of Shackleton, hunting water ice in the shadows next door, carrying payloads from Russia, Egypt, Bahrain, Italy, Switzerland and Thailand. It's a genuinely international mission and it's two weeks out. Watch the investigation.

Four landers, all at the gate

Quieter, and easy to miss under the louder news: on August 4, NASA published a status update on its commercial Moon-base cargo landers, and the striking thing is how many are in final qualification at the same time.

• Blue Origin's Blue Moon MK1 finished an extensive environmental test campaign, including thermal-vacuum at Johnson Space Center, plus communications checkouts against both TDRSS and the Deep Space Network
• Intuitive Machines completed long-range thermal-vacuum testing on IM-3, confirming its sensors read accurately at both ends of the thermal range — and is pairing the lander with the Altus-1 data-relay satellite
• Firefly is building Blue Ghost Mission 2, with its Elytra spacecraft
• Voyager Technologies' Griffin-1 is in environmental testing at JPL, targeting launch late this year

NASA also has three Northrop Grumman technology-demonstration payloads in the mix. Deliveries for this phase are targeted at 2028, inside a program that runs through 2029 and contemplates more than twenty robotic landings.

Thermal-vac and comms checkouts are not exciting to read about. They are, however, the two things that most reliably kill a lander that otherwise works fine: a vehicle that overheats in sunlight and freezes in shadow, and a vehicle nobody can talk to. Four companies clearing those gates in the same season is the clearest signal I've seen that lunar delivery is turning into a supply chain rather than a series of events.

You can watch the whole board on the lunar dashboard.

And on Mars

Curiosity turned fourteen on August 6 — fourteen Earth years on a mission designed for one Mars year — and is still climbing Mount Sharp, closing on the erosional supersurface the team has been driving toward since spring. I wrote about what fourteen years actually costs in yesterday's Sunday Sol.

Two weeks to Chang'e-7. Assuming the rocket cooperates.

— Elio

References

• SpaceX rocket's impact crater on the moon spied by Korean lunar orbiter — Space.com, August 2026
• Danuri captures before-and-after images of Falcon 9 debris impact on moon — Korea JoongAng Daily
• Korean lunar orbiter snaps first pics of the SpaceX Falcon 9 crash site — Engadget
• First images of the site where the Falcon 9 collided with the Moon — Universe Magazine, August 7, 2026
• NASA Will Attempt to Observe Rocket Part's Lunar Impact — NASA, August 4, 2026
• Chinese Long March 7A rocket explodes shortly after liftoff — SpaceNews, Andrew Jones, August 10, 2026
• China says Long March 7A rocket launch failed after flight anomaly — Reuters via CP24, August 10, 2026
• NASA Provides Updates on Moon Base Cargo Landers, Tech Demonstrations — NASA, August 4, 2026
• Upper stage impacting the moon on 2026 August 5 — Project Pluto (Bill Gray)

Forwarded this? Get The Decisional Pass in your inbox: