NASA Safety Panel Seeks Artemis Review Over 22-Launch Refueling Plan

Twenty-two supporting Starship launches may be required before NASA’s planned lunar lander can leave Earth orbit for the Moon. That NASA planning estimate far above SpaceX’s projection of six to eight tanker flights has prompted the Aerospace Safety Advisory Panel to prepare a formal recommendation that the agency reassess its Artemis IV mission architecture.

The panel outlined its concerns during its fourth-quarter public meeting on October 22, 2026. Panel member and former astronaut Charlie Precourt said the issue extends beyond the disputed launch count: Artemis IV’s planned 2028 crewed landing depends on an orbital depot, repeated tanker flights and cryogenic propellant transfers that have not yet matured into a demonstrated operational system. The panel intends to recommend that NASA reconsider the mission’s objectives and architecture to establish a more balanced risk posture.

Why the launch estimate matters

The difference between six to eight flights and approximately 22 is not a minor planning disagreement. It changes the scale of the campaign needed to prepare one lunar lander for departure. Every tanker must launch, reach the intended orbit and support a transfer operation. A larger campaign also places more demand on launch cadence, vehicle turnaround and coordination between the tanker spacecraft, depot and lander.

The count therefore acts as a shorthand for several unresolved engineering assumptions: how much usable propellant each tanker can deliver, how much is consumed getting into position, how effectively fuel can be transferred and how much is lost while stored. The supplied estimates do not resolve those variables, but their wide separation shows that NASA and SpaceX are not yet working from a publicly aligned description of the campaign’s operational scale.

That uncertainty carries a direct schedule consequence. If the architecture requires approximately 22 supporting launches, delays or unsuccessful operations have more opportunities to interrupt the sequence than in a six-to-eight-flight campaign. The safety panel described the repeated, rapid-turnaround transfers as a source of compounding failure modes for a human lunar mission. The concern is architectural because the landing depends on the entire chain working, not merely on the lander performing correctly after it reaches the Moon.

Cryogenic transfer remains a development challenge

Starship uses cryogenic propellants that must remain extremely cold. Storage and transfer place demanding conditions on tank-pressure control, fluid behavior, transfer-line preparation, seals and automated couplers. NASA says cryogenic refueling between two spacecraft has not yet been performed in orbit, although development and testing are advancing.

SpaceX did conduct a tank-to-tank liquid-oxygen transfer during a March 2024 Starship flight. That was useful progress inside one vehicle, but NASA’s own technical guidance says transfer between independent spacecraft has never been demonstrated. An operational lunar campaign requires more than moving fluid once: separate vehicles must rendezvous, connect autonomously, manage ultra-cold propellant and repeat the process at the reliability and pace demanded by the mission plan.

NASA has tested developmental cryocouplers with liquid nitrogen on the ground, including connected and disconnected configurations and simulated docking misalignment. The agency characterizes those devices as early in development, with future campaigns needed to tailor and assess them against specific mission requirements. That gap between basic functionality and repeated flight operations helps explain why the advisory panel is treating refueling as a program-level concern rather than a routine servicing task.

Artemis III reduces some risk, but not the refueling burden

NASA has already changed the sequence leading to the landing. Its revised Artemis III mission, planned for mid-2027, is intended to test rendezvous and docking operations in low Earth orbit with test articles of SpaceX’s Starship and Blue Origin’s Blue Moon lander. The crewed lunar landing has moved to Artemis IV in early 2028.

That Earth-orbit mission can give NASA experience coordinating Orion with two commercially developed lander systems. It can evaluate docking behavior, communications, software interaction and mission-control procedures before astronauts attempt a landing. It does not by itself demonstrate the full tanker sequence, depot operation or repeated cryogenic transfers needed to send the Starship lander toward the Moon.

For NASA, the trade is now between preserving the 2028 landing objective and requiring enough ground and flight evidence to understand the actual risk carried by the refueling architecture. Astronauts depend on that judgment, while taxpayers bear the consequences of schedule changes or additional testing. Until the six-to-eight versus 22-launch gap is narrowed by a validated operating plan and flight demonstrations, the refueling campaign remains a defining constraint on Artemis IV not a supporting detail.

By David Whitaker — Associate editor for AMI’s aerospace and drone systems desk, translating flight systems, aircraft programs, spaceflight, and UAV developments into accessible technical stories.