We’ve observed with concern a rising trend in human spaceflight discussions: the suggestion that we bypass the moon and send humans directly to Mars. In light of this, we believe it is critical to identify and address the risks inherent in a “direct to Mars” approach. Equally important is articulating why a “moon first, then Mars” pathway offers a lower-risk, more strategic and deliberate trajectory, both from a human safety and overall mission success perspective.
First, let us make clear that a human mission to Mars is both technically achievable and scientifically compelling. NASA’s official reasons include science as well as national posture and inspiration among the reasons to go. In our search for life elsewhere in the universe, Mars is the closest and most promising candidate. Exploring Mars can help us understand the broader universe, planetary processes and our own planet’s evolution and ecosystem. Take for example how the past 20 years of robotic exploration have provided important clues about the possibility of ancient life, and may yet reveal the fingerprints of a former biosphere. But finding definitive evidence of past or extant life will likely require humans on the surface. And if we find evidence, human investigators will be essential to determine whether that life originated on Earth or arose independently. Pursuing these questions lies at the heart of astrobiology, the discipline that merges biology and planetary science to understand the origins, evolution, and distribution of life in the universe.
While we’ve learned a great amount from our robotic exploration to date, the tasks required to characterize Martian geology in sufficient detail, and to collect the right samples for return and analysis, are well beyond the capabilities of current robotic systems. Humans on the ground can accomplish in a single day what robotic spacecraft take a year to do — and with far greater fidelity, underscoring the need to send humans to Mars.
We also have a bit of history flying humans in space, whether in low Earth orbit (LEO), on the moon, or through analog studies on Earth. However despite this impressive experience base, many key milestones on the path to Mars remain outside our current capability and understanding. While sending humans to Mars is achievable today, it carries significant risks, many of which are not always apparent.
Consider that when Apollo went to the moon, the entire Mercury and Gemini programs were designed to de-risk human spaceflight, demonstrating step-by-step our ability to meet key technical milestones. Some may be tempted to view our success in LEO, particularly with the International Space Station (ISS), as that final stepping stone before we journey to Mars. But it’s important to recognize just how many critical ‘firsts’ we would be attempting — all in a single mission — if we bypassed Artemis and went directly to Mars without an intermediate campaign to the moon that would allow us to reduce technical and human risk. Some of these must-succeed ‘firsts’ include:
- The longest continuous human mission in space: The current record for continuous time spent in the space-mission environment is 438 days. The first crewed Mars mission with a surface excursion will likely exceed 730 days. In addition to the physiological effects of space travel, the extended isolation from friends, family and Earth itself, amplified by the loss of real-time communication, will be unlike anything any human has ever experienced.
- Unprecedented radiation exposure: This will be the first time humans are exposed to 1 Sievert of cumulative space radiation during a single mission. For context, this is roughly equivalent to 10,000 chest x-rays per astronaut. However, the type and dose-rate of radiation in deep space is very different and far less understood, which means this comparison may significantly underestimate the true risk.
- First time without real-time support from Mission Control: For the first time, astronauts will have to operate independently without the constant oversight of roughly 80 experts in Mission Control Centers (MCC) who monitor, identify, diagnose and respond to anomalies aboard the ISS. The crew will be responsible for identifying and resolving issues on their own, with communication delays up to 45 minutes round-trip. And when Earth and Mars are on opposite sides of the Sun, they could be completely out of contact with Earth for up to two weeks.
- First reliance on pre-staged assets and in-situ resource utilization: This will be the first time astronauts must survive using systems and supplies either delivered in advance or produced on Mars. Keep in mind that food and pharmaceuticals cannot be effectively pre-positioned due to degradation over time and must travel with the crew. It will also be the first time astronauts will eat solely pre-packaged food without access to fresh food for more than three months, and to date, we do not have a food system that can support a round-trip Mars mission.
- First time managing severe injury or death in deep space: Mars astronauts must be prepared to face serious illness, injury or death of a teammate, both emotionally and operationally. They must do this without real-time support of MCC. This includes adapting to the loss of key roles, tasks and mission objectives as the remaining crew works to complete the mission.
- First self-recovery after landing in partial gravity: This will be the first time astronauts must self-extricate from a landing vehicle and begin functioning without a ground support team. What’s not widely publicized is that, after much shorter missions in LEO, astronauts typically rely on recovery crews of 20 or more people to assist with their inability to walk and with significant nausea, vomiting and disorientation that occur as they re-adapt to gravity. We currently have no way to assess how difficult that will be to do in Mars gravity.
- First extravehicular activity (EVA) without real-time support: Astronauts on Mars will perform far more EVAs than ever attempted, over longer durations, and without direct communication with Mission Control. This will place unprecedented physical and operational demands on both the crew and the spacesuits.
- First planetary ascent by a human crew: For the first time, astronauts must launch themselves from the surface of another planet without the support of a ground crew. This means a small crew will perform all infrastructure tasks, vehicle preparation, fueling, troubleshooting, and liftoff. This is a dramatically different scenario than the Apollo missions.
Some ‘firsts’ cannot be attempted until we go to Mars. Entry, descent and landing for a human mission requires the ability to safely land approximately 40 metric tons of material through Mars’ thin atmosphere. This is unlike landing on Earth or at the moon, and to date, our robotic Mars missions have only demonstrated the ability to land roughly one metric ton.
However, many other ‘firsts’ can be demonstrated at the moon and in cis-lunar space through the Artemis campaign — but only if those missions are purposefully designed to gather the right data and test the right systems. This is the intent of NASA’s Moon to Mars Architecture, to demonstrate key milestones on our road to Mars, just as Mercury and Gemini paved the way for the Apollo missions to the moon.
Our goal should be to minimize the number of critical ‘firsts’ that our astronauts are attempting in the first Mars mission because on the moon, we have a reasonable chance of bringing our astronauts home if something goes wrong. We have a margin for error on the moon, and history has shown that we need that margin. Rarely do we get everything right on the first attempt.
Lunar space also allows us to simulate Mars-like conditions. We can intentionally introduce time-delayed or lost communications, retaining the ability to restore communications in an emergency. And if necessary, we have the potential to bring an ill or injured astronaut back to Earth in three to 11 days from lunar space. After a Mars transfer burn, no such return or rescue is possible.
Trying new approaches and solutions on the moon first gives us a much higher tolerance for imperfection in design, engineering, logistics and implementation. It allows for learning as we go. Attempting to conduct field geology on another world for the first time, while using newly designed tools and space suits, under the pressure of global attention, is daunting. On the moon, we are not as constrained by mission duration, and we have the opportunity to make mistakes.
This same need for programmatic tolerance also applies to the transit phases of a Mars mission. Astronauts will likely spend at least 18 months inside a transit vehicle while experiencing the compounding effects of being deconditioned by microgravity, radiation, isolation, confinement and sustained operational stress. Even with decades of human spaceflight experience, we still have only a nascent understanding of how the human mind and body respond to deep space conditions over long duration missions. Our data effectively ends around six months with all the comforts of the ISS. Gateway is the only proposed program that allows us to study these full-body multi-system deconditioning effects over extended timeframes. ISS remains safely inside the Earth’s magnetic sphere, which means that Gateway is our only meaningful analog for the transit phases of a Mars mission, and it is expected that much of its hardware will directly inform the design of the Mars transit vehicle.
In his recent testimony to Congress, NASA-Administrator-designate Jared Isaacman indicated that we could do the moon and Mars in parallel at the same time. While not entirely simultaneous, since we’re nearly ready to go back to the moon now, we can begin planning for Mars while conducting missions in lunar space. By the time we’re ready to go to Mars, our lunar activities will be mature, enabling a programmatic shift towards Mars as a first priority.
Regardless of which combination of government agencies and commercial company capabilities leads to the first attempt at sending humans to Mars, the technical challenges will remain the same. Sending humans to Mars will serve to marshall the best of our capabilities in fields like engineering, computation, biology, medicine, human performance and more. The first astronauts who do eventually journey to Mars will face many unknowns about how their bodies and minds have been affected when they return. There is no way around that. But we can significantly reduce the risks associated with the first Mars mission by intelligently using the moon as a stepping stone.
We’ve come so far, step by step, to reach the point where interplanetary travel is now a real possibility. Let’s continue taking those steps to Mars carefully and deliberately, rather than attempting a leap too far.
Erik Antonsen was Assistant Director of Human Health and Performance for Human System Risk Management at NASA Johnson Space Center and Element Scientist for Exploration Medical Capabilities for the Human Research Program.
Jennifer Rochlis was the Division Chief for Human Systems Engineering and Integration at NASA Johnson Space Center and has worked extensively in both the private and government sector, focusing on human integration with complex systems.
Bruce Jakosky has been a Mars researcher for almost 50 years and was the Principal Investigator for the MAVEN Mission that explored the Martian upper atmosphere.
Scott Hubbard served as the first Mars Exploration Program Director, created NASA’s Astrobiology Institute, was Center Director for NASA Ames, and founded the Stanford Center for Commercial Space Transportation.
This article first appeared in the May 2025 issue of SpaceNews Magazine.
SpaceNews is committed to publishing our community’s diverse perspectives. Whether you’re an academic, executive, engineer or even just a concerned citizen of the cosmos, send your arguments and viewpoints to opinion@spacenews.com to be considered for publication online or in our next magazine. The perspectives shared in these op-eds are solely those of the authors.
