SpaceX states that two major Starship Human Landing System tests are targeted for 2026: a long-duration orbital flight and the first ship-to-ship propellant-transfer demonstration. The sequence begins with one Starship remaining in orbit to measure propulsion, thermal behaviour, propellant storage, and boil-off; a second vehicle will then rendezvous and transfer propellant. NASA selected Starship as the Artemis III lander in 2021 and later selected it for Artemis IV.
Technically, orbital refuelling is the enabling capability for the Artemis lunar architecture and could allow Starship to deliver up to 100 tons directly to the lunar surface. That payload class could support large habitats, rovers, power systems, excavation equipment, and redundancy essential to civilisation preservation. However, long-duration cryogenic storage, boil-off, rendezvous reliability, landing precision, and rapid reuse remain major unproven dependencies; failure in any one can delay or constrain permanent settlement.
The Ark team should track the 2026 flight campaign against measurable gates: successful long-duration storage, quantified propellant boil-off, autonomous rendezvous, transfer efficiency, engine relight performance, lunar-landing accuracy, and delivered mass. Ark architecture studies should model both a 100-ton nominal cargo stream and substantially lower-throughput contingencies, while prioritising payloads that reduce dependence on repeated Earth launches, including power generation, closed-loop life support, radiation shielding, manufacturing, and archival storage.