When Will Humans Reach Mars?

When Will Humans Really Go to Mars? What Has to Happen First

A 2026 reality check on NASA, SpaceX, Starship, radiation, life support, Mars landing technology - and the chain of milestones that must work before the first crew can leave Earth.

Mars does something strange to calendars. A new rocket flies, NASA publishes a new architecture, SpaceX moves another Starship milestone forward - and suddenly the first human landing feels close enough to circle on a date. Then the hard parts reappear: a life-support system needs years of testing, orbital refueling is still unproven at the required scale, or engineers return to the stubborn fact that landing a very heavy spacecraft on Mars is unlike anything we have done before.

So when will humans actually go? In 2026, the most useful answer is not a year. It is a sequence. If the major pieces mature with relatively few surprises, a crewed mission in the 2030s remains possible. If even a couple of the hardest systems need extra development cycles, the first surface expedition can move into the 2040s very quickly.

That is less satisfying than a countdown clock, but it is closer to how Mars will really be reached. The problem is no longer simply sending a spacecraft toward another planet. It is sending a transport system, a habitat, a medical clinic, a power plant, a communications network and a return plan - then expecting all of them to keep working far beyond any practical rescue.

Getting humans to Mars is not just a question of building a powerful rocket. The mission also requires reliable propulsion, radiation protection, life-support systems, heavy-payload landing technology and a way to bring the crew home.

Mars Is Six Months Away - and 26 Months Away

On an energy-efficient trajectory, Earth-to-Mars transit is usually discussed in terms of roughly six to nine months. The more important number, however, is 26 months. That is about how often Earth and Mars return to a favorable geometry for relatively efficient transfers. Miss the window and a program cannot simply launch a few weeks later. The next good opportunity is more than two years away.

That rhythm shapes every serious mission plan. Cargo is likely to go first: power hardware, habitats, spare parts, communications equipment, surface vehicles and perhaps equipment for making oxygen or propellant. A crewed Mars mission will probably look less like one giant Apollo-style launch and more like a campaign assembled across several planetary launch windows.

The first human Mars mission will probably be a campaign, not a launch.

NASA studies have therefore treated a round trip as a mission measured in years. A conjunction-class expedition may keep astronauts away from Earth for two years or longer because the return trajectory can require a long stay near Mars. Faster propulsion can shorten the cruise, but it cannot make orbital mechanics disappear.

What NASA Actually Says in 2026

NASA still describes Mars as the long-range destination of its human exploration program and says technologies now in development could support missions there in the 2030s. That is not the same as having a booked launch date. The current Moon to Mars architecture is deliberately stepwise: use the Moon and deep-space operations to prove capabilities, then carry the lessons into a Mars campaign.

That is an important distinction. NASA is not working toward a public Mars landing date in the way Apollo once worked toward the end of the 1960s. The agency is working through a list of risks: deep-space habitation, human health, communications, autonomy, surface power, entry and landing, propulsion, logistics and the ability to operate for long periods without immediate resupply.

The Moon is part of that process, not a detour from it. Artemis II flew around the Moon in April 2026, and later Artemis missions are intended to push crews farther from the quick-return safety of low Earth orbit. NASA's architecture work is also starting to treat Mars communications and navigation as systems that have to be designed, not assumed.

SpaceX Is Working on a Much Faster Clock

SpaceX is following a much more aggressive schedule. Its Starship program is built around full reusability, very large payloads and refueling in orbit, with Mars as an explicit long-term destination. The company currently points to cargo missions to the Martian surface no earlier than 2028 and has also discussed a future roughly two-year crewed Mars flyby mission. Neither is a firm launch commitment.

The difficulty is that a Mars-capable Starship is not just a bigger launch vehicle. The full chain still has to work: repeatable orbital operations, high flight cadence, large-scale cryogenic refueling in orbit, long-duration spacecraft reliability, crew systems, Mars entry, controlled landing, surface operations and - eventually - departure from Mars.

That is why framing Mars as NASA versus SpaceX is not especially useful. NASA increasingly depends on commercial transportation, while SpaceX is running into the same human-health, landing and surface-system problems that government research programs have studied for decades. In practice, the paths overlap.

A conventional Mars transfer can keep astronauts in deep space for many months. During that time, the spacecraft must function as a home, laboratory, hospital and life-support system with very limited possibility of outside help.

The Hard Part Starts After the Rocket Leaves Earth

A rocket powerful enough to send people toward Mars is only the opening move. The mission becomes truly difficult once the crew is months from home, Earth is a delayed radio conversation away, and a serious failure can no longer be solved by launching a spare part or bringing the astronauts back.

Recent research does not point to one impossible 'Mars problem.' It points to many difficult problems that have to be made acceptably safe at the same time. That is both encouraging and uncomfortable. A mission can survive one imperfect subsystem. It cannot survive a long chain of systems that are all barely good enough.

1. The Crew Has to Survive Deep Space Before It Reaches Mars

Beyond Earth's magnetic protection, astronauts are exposed to galactic cosmic rays and solar energetic particles. A 2025 npj Microgravity study used Artemis I measurements to validate radiation-exposure predictions along deep-space trajectories relevant to Mars. Radiation is especially awkward because the most energetic particles are difficult to stop, while every extra layer of shielding adds mass.

For the biology in more detail, see our article on how the human body may change on Mars. For mission design, the key point is that these risks interact. A crew does not experience radiation, bone loss, sleep disruption and isolation as separate laboratory experiments; it experiences all of them together.

The risks do not arrive one at a time. Microgravity, isolation, disrupted sleep, immune changes, vision problems, muscle and bone loss, and possible neurological effects all overlap during a mission that may last years. A 2025 Nature Reviews Neuroscience review, for example, examined how microgravity and cosmic radiation may affect the central nervous system - the same system the crew depends on for judgment when something goes wrong.

Mars also breaks the medical model used on the International Space Station. On the ISS, a badly injured astronaut can in principle return to Earth. A Mars crew cannot. Reviews published in Acta Astronautica in 2025 therefore treat medical autonomy - onboard diagnostics, treatment capability and decision support - as a mission requirement rather than a convenience.

2. The Ship Has to Work Like a Tiny World

The International Space Station lives inside a supply chain. Cargo vehicles arrive, replacement hardware can be launched, and specialists on Earth can help troubleshoot almost in real time. A Mars spacecraft will have to live without most of that safety net.

A 2026 review of Environmental Control and Life Support Systems examined more than 270 papers and technical reports. Its message was unsurprising but severe: long missions need systems that are more closed-loop, autonomous and repairable than today's orbital hardware. Water has to be recovered repeatedly. Air has to be regenerated. Waste becomes feedstock. Food is not just food; it is part of the life-support architecture.

NASA is probing the human side of the problem through CHAPEA. The second 378-day simulated Mars mission runs through October 2026 inside a 1,700-square-foot habitat in Houston. The crew works under resource limits, equipment failures, crop cultivation, heavy workloads and communication delays that can approach 22 minutes one way. In May 2026, NASA also imposed a simulated two-week communications blackout.

Houston is not Mars, and nobody pretends it is. The value of an analog is simpler: find the overlooked failure while the habitat door can still be opened from the outside.

3. Landing 20 Tonnes Is Nothing Like Landing a Rover

Mars has an inconvenient atmosphere. It is thick enough to generate intense heating during entry, yet too thin for a very heavy human-rated lander to rely on parachutes alone. Perseverance, one of the heaviest objects ever placed safely on Mars, arrived with a landed mass of roughly one tonne. Human mission studies often work with payloads around 20 tonnes or more. That is not an incremental increase; it is a different landing problem.

NASA's Entry, Descent and Landing Architecture Study examined systems for roughly 20-tonne human-class payloads. Conventional parachutes do not scale neatly to that regime, which is why supersonic retropropulsion is so important: large engines may have to fire while the vehicle is still moving faster than sound through the Martian atmosphere.

SpaceX's experience returning Falcon boosters and developing Starship is relevant here, but it is not a shortcut to a crew-rated Mars landing. Mars brings different atmospheric density, gravity, dust, terrain, communications delays and, on the first missions, essentially no local infrastructure.

Mars is thick enough to burn a spacecraft — and thin enough that parachutes alone cannot save a very heavy one.

 

Mars has enough atmosphere to create dangerous heating during entry, but not enough to slow a massive crewed spacecraft with parachutes alone. Landing tens of tonnes safely remains one of the central engineering challenges of a human Mars mission.

4. The Return Trip Has to Exist Before the Crew Leaves Earth

Designing a one-way mission is easier. That is exactly why credible crewed architectures do not treat the return journey as something to solve later.

Before astronauts depart, planners will want unusually high confidence that the hardware needed to survive and leave Mars is already in place or can be created reliably on the surface. That may mean pre-deployed habitats, power systems, communications assets, spare parts, ascent hardware and equipment for producing oxygen or propellant.

MOXIE, the small experiment carried by Perseverance, proved one useful point. Across 16 runs it extracted 122 grams of oxygen from Mars' carbon-dioxide-rich atmosphere and reached production rates of up to 12 grams per hour. That is nowhere near the industrial scale a crew would need, but it showed that making a critical resource on Mars is an engineering problem, not pure speculation.

Scaling that idea is the hard part. Starship's Mars concept depends on methane and oxygen, with both ultimately envisioned as products of Martian resources. No one has yet run a full-scale propellant plant on Mars, stored large quantities of cryogenic propellant through the local environment and then launched a human-rated vehicle from the surface. Each step is plausible on paper. The complete chain is not yet demonstrated.

5. Mars Needs Power on Its Worst Day

A crewed outpost cannot become uninhabitable because dust reduced solar output or a battery bank failed at the wrong time. That is why NASA continues to study surface fission power for Mars. Life support, heating, communications, science equipment, workshops, water processing and any serious propellant-production system all need electricity when conditions are least convenient.

Power also buys options. More dependable electricity means more recycling, more heating margin, more water processing, better workshops, stronger communications and a larger buffer when another system fails. A Mars expedition is as much an energy project as a transportation project.

A first Mars expedition will need far more than a lander. Power generation, communications, habitats, mobility, water processing and emergency supplies may need to be deployed and tested before astronauts ever leave Earth.

So When Could the First Human Mission Actually Happen?

There is no honest way to turn this stack of engineering problems into an exact launch date. What we can do is look at the milestones visible in 2026 and ask what each decade would require.

2026-2030: Prove the Transport System

For Mars, Starship has to grow from a flight-test program into a dependable transportation system. Reaching orbit is only one part of that. The more consequential milestones are rapid reuse, consistent heat-shield performance, orbital refueling, long-duration propellant management and the ability to move very large cargo loads repeatedly.

SpaceX currently points to Mars cargo missions no earlier than 2028. That should be read as a target, not a promise. Even an uncrewed landing attempt would be enormously valuable because it would replace some of today's simulation with real data from a Mars-scale vehicle.

NASA, meanwhile, will keep using Artemis, low Earth orbit, CHAPEA and other analog programs to reduce human-system risk. A 2026 Acta Astronautica paper argues that the ISS and future commercial stations could also serve as Mars analog platforms, letting crews practice greater autonomy, delayed communications, food systems, medical care and countermeasures while Earth is still close enough to provide an escape route.

Early 2030s: Possible, but Everything Has to Go Right

Could people leave for Mars in the early 2030s? Physics does not forbid it, and NASA continues to discuss technologies that could support human Mars missions in that decade. A surface expedition that early, however, would require an unusually clean run of successes across heavy launch, orbital refueling, deep-space crew systems, Mars landing, surface power and return capability.

A flyby or orbital mission could plausibly come first because it avoids the most dangerous step: landing a large crewed vehicle and then getting it off Mars again. SpaceX has discussed a future crewed Mars flyby profile for that reason, although no firm launch date is attached to it.

Late 2030s: A More Credible Window

The late 2030s appears often enough in NASA technical studies to be more than a science-fiction date, but it should not be mistaken for an official landing commitment. Some transportation and propulsion studies use late-2030s Mars missions as design cases precisely because the systems involved still need a long maturation period.

In other words, the late 2030s is a useful engineering horizon: close enough to design toward, far enough away that several generations of hardware can still fail, improve and fly again.

The 2040s: Still a Plausible Outcome

If orbital refueling takes longer than expected, if human-scale Mars landing needs several robotic precursors, if radiation countermeasures remain uncertain, or if lunar programs consume more time and funding, the first surface expedition can slip into the 2040s without any single dramatic failure.

Mars makes delays compound. Launch opportunities arrive about every 26 months. Miss two of them and a program can lose more than four years even before additional development time is counted. That is one reason confident Mars dates age so badly.

What Could Make Mars Happen Faster?

The biggest accelerator may not be a revolutionary engine. It may simply be the ability to test complete systems much more often.

If reusable heavy launch makes hundreds of tonnes to orbit affordable and routine, mission design changes. Engineers can fly more shielding, more spares and larger habitats. Cargo missions can go ahead of the crew. Redundancy becomes something you can buy with mass instead of something you have to optimize away.

That is the deeper reason Starship matters even if its schedule moves. The promise is not merely a very large spacecraft. It is a different mass budget for human exploration.

Other accelerators could include better nuclear propulsion, more capable autonomous medical systems, robust closed-loop life support, surface resource utilization and robotic construction. None removes the need for testing, but all become more useful if launch mass and launch frequency are less punishing constraints.

What Could Push the Mission Back by a Decade?

Several delays that look small from Earth become large on a Mars calendar.

Human-rating takes time. A prototype that succeeds nine flights out of ten may be impressive; it is not good enough as a lifeboat. Mars landing offers little room for trial and error with a crew onboard. The return system has to be trusted before departure. Medical uncertainty grows with mission duration. And budgets, industrial capacity and program priorities can change much faster than orbital mechanics.

Integration may be the least glamorous risk and one of the most dangerous. Every subsystem can pass its own test and the mission can still fail because power margins, maintenance schedules, software, interfaces or crew workload collide in ways no individual test revealed. That is why long-duration integrated tests matter so much.

The First Mars Mission Probably Will Not Look Like The Martian

The first astronauts on Mars are unlikely to step into a comfortable settlement with greenhouse windows and a fleet of shiny rovers. A first expedition will probably look more like an isolated industrial field station: compact, heavily engineered and designed around survival rather than spectacle.

Cargo may already be waiting. Robots may have prepared landing zones or moved equipment. Power systems could be running before the crew leaves Earth. The astronauts may spend far more time maintaining hardware than exploring, and every excursion outside the habitat will consume resources, suit life and risk margin.

The psychological shift may be just as important. Near maximum separation, a radio message can take more than 20 minutes to travel one way. A simple question and answer can take roughly three quarters of an hour. Earth will still support the crew, but it will no longer function as mission control in the Apollo sense. On Mars, many decisions will belong to the people who are there.

On a Mars mission, astronauts may have to diagnose medical problems, repair critical systems and manage radiation exposure without immediate help from Earth. Autonomy becomes as important as propulsion.

The Date Matters Less Than the Sequence

Humans will go to Mars when the mission no longer depends on a string of heroic assumptions.

A reusable heavy-lift system has to become dependable. Orbital refueling has to work routinely. A deep-space habitat has to keep people healthy for months. Radiation risk has to be acceptable. Human-scale payloads have to land precisely. Surface power has to survive bad conditions. Water and oxygen loops have to keep running with little resupply. The crew needs medical autonomy. And the route home has to exist before anyone boards the spacecraft on Earth.

None of those requirements is fantasy in 2026. That is why Mars genuinely feels closer than it did twenty years ago. There is hardware, flight data, laboratory work, analog missions and serious engineering behind almost every part of the architecture.

But 'almost every part' is doing a lot of work in that sentence. The early 2030s remains an aggressive possibility. The late 2030s is easier to defend as a planning horizon. The 2040s would not be surprising if the hardest systems need more iterations than their developers hope.

The first footprint on Mars will not appear because a billionaire, an agency or a single rocket finally declares the moment has arrived. It will appear when dozens of mostly unglamorous systems become reliable enough that sending people is no longer a stunt - and bringing them home is part of the design from the beginning.

FAQ: Human Missions to Mars

When will humans first go to Mars?

NASA is developing technologies for human Mars missions in the 2030s, but there is no firm landing date. An early-2030s surface mission would require an unusually fast run of successful development; the late 2030s or 2040s remain credible windows depending on technology readiness.

How long does it take to fly to Mars?

On conventional energy-efficient trajectories, a one-way trip is usually estimated at roughly six to nine months. Faster propulsion could shorten the cruise, but at the cost of greater technical complexity and energy demand.

Why can we only launch to Mars every two years?

Earth and Mars return to favorable transfer geometry about once every 26 months. Missing one launch opportunity can therefore push a mission by more than two years.

What is the hardest part of sending humans to Mars?

There is no single hardest problem. The mission has to solve long-duration health and radiation, closed-loop life support, heavy Mars landing, dependable surface power, medical autonomy and a credible way to leave Mars and return to Earth.

Will SpaceX or NASA send humans to Mars first?

There is not enough evidence to say. NASA is building a broader Moon-to-Mars exploration architecture, while SpaceX is developing Starship as a reusable high-capacity transport system. Their approaches increasingly overlap, and a future mission may depend on technologies developed by both public and commercial programs.