The Mars Contest

Not terrain

Nothing on Mars Is Scarce Enough to Fight Over

The lunar contest is about real estate. The Martian one cannot be, and that changes its shape entirely.

At the lunar south pole the number of locations with both near-continuous illumination and short access to shadowed volatiles is plausibly in the single digits, which makes it a genuine allocation problem with two programs walking toward the same ridge. Mars has the land area of Earth’s continents, eight strong candidate regions on the shortlist alone, and no equivalent of the peak of eternal light. Two programs could land 2,000 km apart and never interact.

So the competition runs on other axes. Who demonstrates the capabilities that a crewed architecture depends on. Who returns the first Martian material and gets to make the first claim about Martian biology. Whose operating norms — particularly on contamination — become the de facto standard because they were used first. Those are contests about precedent and prestige rather than about resources, and they resolve on a different timescale than a territorial dispute would.

One caveat keeps this from being too tidy. If Phobos turns out to hold accessible water, the orbital node is scarce — one small body, a handful of viable anchoring sites, and everyone’s architecture wanting the same rock. MMX answers that in 2029, and a positive answer converts part of the Mars contest into something much more like the lunar one.

The programs

Who Is Actually Flying Hardware to Mars

Two nations have landed and operated on the surface. Everyone else has orbiters or plans.

One line summarizes the strategic position: no actor has a funded end-to-end crewed Mars architecture. NASA has a pathway routed through the Moon, China has a published roadmap with hardware only at the robotic end, and SpaceX has a vehicle and a slipping schedule. The crewed contest is currently a contest of intentions.

Ascent

The First Launch From Mars Will Be Robotic, and Probably Chinese

Every orbit-staged crewed plan depends on two operations nobody has performed. One mission is funded to perform both.

Every architecture in which the transit vehicle stays in orbit requires launching from the Martian surface and rendezvousing with something in Mars orbit. Neither has ever been done. Both are on Tianwen-3’s critical path, because a sample return mission must lift its sample off the surface and hand it to a return vehicle.

The capability that unlocks crewed orbit-staging arrives on a robotic science mission, and it arrives first for the program that is not publicly racing for it. That is a genuine strategic asymmetry and it is rarely stated: the demonstration matters far more than the samples it delivers, because ascent and rendezvous are exactly the pieces the crewed architectures are missing.

The counterweight is that neither the ascent vehicle nor the rendezvous hardware scales. A sample canister is kilograms; a crew ferry is tens of tonnes with an aeroshell, life support and human rating. Demonstrating the operation is not the same as owning the vehicle, and the crewed descent-and-ascent vehicle remains unfunded everywhere — the master gate described on Mission Architectures.

Samples

Sample Return Primacy Is in Play, and It Was Not Supposed to Be

Perseverance has been caching samples since 2021. The mission to collect them is the one in trouble.

NASA and ESA’s Mars Sample Return was the flagship: Perseverance fills tubes, a later mission retrieves them, an orbiter carries them home. Perseverance has done its part. The retrieval architecture ran into cost and schedule findings severe enough to force a restructuring and a competitive search for cheaper approaches, and the program’s return date has moved repeatedly.

Tianwen-3 is designed to collect and return its own samples on a single simpler mission, targeting the 2028 and 2031 windows. If it flies as planned, the first Martian material on Earth will be Chinese, and the first published claim about Martian biosignatures will be made from it.

Why that matters beyond prestige: the sample answers the astrobiology question, and the astrobiology answer sets the planetary protection regime for everything afterward. Whoever holds the first definitive result on Martian life shapes how restrictive the rules become — and those rules bind crewed programs and settlement plans directly, as set out on Martian Water & Ice.

Cargo

The Mass Asymmetry Runs the Other Way

One vehicle in existence or development can put a hundred tonnes on Mars, and it is American and commercial.

The robotic scoreboard favors China on near-term firsts. The cargo scoreboard does not come close. A crewed Mars program needs roughly 230 t of landed hardware for a minimum outpost and far more for anything beyond it, which means vehicles in the 100 t landed class. Starship is the only vehicle in existence or serious development in that class anywhere.

Long March 10 is a lunar-class launcher — capable, human-rated in design intent, and not a Mars cargo vehicle. A Chinese crewed Mars campaign on the announced orbit-first architecture would need either a substantially larger vehicle or a great many more flights, and neither has been announced. China leads on the demonstrations and the United States leads on the tonnage, and neither lead converts into a crewed landing without the other side of the ledger.

That asymmetry also explains the different failure modes. An American program fails by schedule slip — the vehicle exists and the dates move. A Chinese program fails by mass — the roadmap exists and the lift does not.

Protection

Planetary Protection Is the Real Governance Fight, and It Is Voluntary

COSPAR policy is a scientific consensus, not a treaty, and it has never been tested against a crewed landing.

The rules that restrict where spacecraft may go on Mars — Special Regions, sterilization requirements, restricted-return handling — come from COSPAR, an international scientific body. Agencies follow them because they choose to, and because their own science communities insist. There is no enforcement mechanism, no penalty, and no court.

A crewed mission breaks the framework by construction. A habitat is a warm, humid, leaking reservoir of terrestrial microbes sitting on a planet whose subsurface ice is the archetypal Special Region, and the crew is going there specifically to mine that ice. No crewed architecture has an approved planetary protection plan, because no regime exists that could approve one.

The enforcement path that does exist runs through national licensing. The Outer Space Treaty makes states responsible for the activities of their non-governmental entities and requires authorization and continuing supervision, so a commercial Martian landing is legally the licensing state’s problem. In practice no state has a statutory mechanism for licensing the planetary protection compliance of a crewed Mars landing, and the first actor to land one will establish the precedent by doing it. That is a governance gap with a date attached.

Law

What the Treaty Actually Says, and What It Leaves Open

Four articles do most of the work, and none of them was written with a settlement in mind.

Two competing frameworks have grown up alongside the treaty, both written for the Moon and both setting precedent that will carry to Mars. The Artemis Accords, now with a large and growing signatory list, introduce safety zones and resource-use norms. The International Lunar Research Station partnership assembles a different coalition around different assumptions. Neither is a treaty, both are coalitions of the willing, and whichever set of norms gets exercised first at Mars becomes the reference the other has to argue against.

Gaps

What Each Program Is Actually Short Of

The shortfalls are specific, different from each other, and none is a technology nobody understands.

Read across the right-hand column and the same two items recur: a crew ferry and a surface reactor. Neither is a research problem. Both are procurement decisions nobody has made, and whichever actor makes them first converts an intention into a program.

What would overturn this

The Events That Would Reshape the Contest

Three of these could happen before 2032.

Record

Provenance and Data Sources