You need to know if the ice is under one meter or ten meters of debris
Humanity's long search for water on Mars has always been constrained by distance — satellites can sense what lies beneath the surface but cannot resolve the depth or structure that would make a drilling mission viable. Researchers at the University of Arizona have now demonstrated that small drones carrying ground-penetrating radar, tested over buried glaciers in Alaska and Wyoming, can provide precisely that missing layer of knowledge. The work positions aerial scouting as a bridge between broad orbital surveys and the moment a future astronaut or rover must commit to breaking ground, transforming a promising scientific hunch into an actionable target.
- Orbital radar has long been able to detect buried Martian ice, but it cannot tell mission planners whether that ice sits one meter or ten meters down — a gap that could doom a drilling effort before it begins.
- University of Arizona doctoral researcher Roberto Aguilar and his team flew drone-mounted ground-penetrating radar over debris-covered glaciers, environments that closely mimic what Mars missions would encounter, and the results were validated by physical excavation.
- The drones resolved thin debris layers, precise ice depths, and internal glacial structures that orbital instruments like SHARAD on the Mars Reconnaissance Orbiter simply cannot distinguish at their scale.
- The team worked through the operational specifics — flight altitude, speed, radar orientation — turning an experimental concept into a deployable methodology ready to inform real mission planning.
- The stakes extend beyond science: water ice on Mars represents drinking water, oxygen, and potential evidence of past life, meaning better targeting tools could determine the success or failure of human survival on another world.
The search for water on Mars has always been a problem of resolution. Orbiting spacecraft can detect ice buried beneath the Martian surface, but they cannot reveal how deep it lies or what overlies it — details that are not academic when a mission must decide where to drill. A team at the University of Arizona, led by doctoral researcher Roberto Aguilar, has demonstrated a practical solution: small drones equipped with ground-penetrating radar, flying close enough to the surface to see what satellites cannot.
To test the concept, the team flew the system over debris-covered glaciers in Alaska and Wyoming — environments that closely resemble the ice deposits identified on Mars. The drones mapped ice thickness with precision, detected debris layers just a few feet thick, and revealed internal structures within the ice. Field excavations confirmed what the radar showed. The team also worked through the operational details that separate a promising idea from a usable tool: the right altitude, speed, direction of flight, and radar orientation relative to the ice.
On Mars, the approach would function as a middle layer in a three-stage strategy. Orbiters would identify broad regions of likely ice. Drones would then scout those regions at high resolution, mapping accessibility before any surface mission commits to a landing site. The concept builds on the precedent set by NASA's Ingenuity helicopter, which proved powered flight is possible in Mars' thin atmosphere.
The stakes are considerable. Water ice on Mars is a record of the planet's climate history, a potential repository of evidence for past life, and a practical resource — drinking water, oxygen, even agriculture — for future human missions. By adapting techniques developed for studying Earth's glaciers, researchers are making the hunt for buried Martian ice something far more precise, and far more survivable.
The search for water on Mars has always been a problem of resolution. Satellites orbiting overhead can spot ice buried beneath the Martian surface, but they cannot tell you how deep it lies or what sits on top of it. For a future mission planning where to drill, that difference matters enormously. A team at the University of Arizona has now demonstrated a solution: small drones equipped with radar, flying low enough to see what orbiters cannot.
Roberto Aguilar, a doctoral researcher leading the work, put the problem plainly. If you want to know whether the ice you are targeting is a meter down or ten meters down, you need a tool that can measure from close range. Orbital radar instruments like SHARAD, which has been mapping Mars from the Mars Reconnaissance Orbiter for years, excel at finding large deposits of subsurface ice across the planet's mid-latitudes. But they lack the precision to resolve the finer architecture—the thickness of overlying debris, the exact depth of the ice, the internal structure of the deposit. These details are not academic. They determine whether a deposit is reachable with the drilling equipment a mission can carry.
To test whether drone-mounted ground-penetrating radar could fill that gap, Aguilar's team flew the system over glaciers in Alaska and Wyoming. These ice masses, buried under layers of rock and debris, resemble the ice deposits scientists have identified on Mars. The results were striking. The drones mapped ice thickness with precision, detected debris layers just a few feet thick, and revealed internal structures within the ice itself. Field excavations and drilling validated what the radar showed. The team also ran simulations to confirm the radar signals were actually coming from beneath the surface debris, not bouncing off something else.
The work required learning the practical details of drone-based radar operation. At what altitude should the drone fly? How fast? In which direction relative to the glacier's flow? How should the radar be oriented to catch the ice? These questions sound mundane, but they are the difference between a promising idea and a deployable tool. Aguilar and his colleagues worked through them systematically, turning theory into operational knowledge.
On Mars, the approach would work as a middle layer in a three-stage exploration strategy. Orbiters would identify broad regions where ice likely exists. Drones would then scout those regions at high resolution, mapping exactly where the ice lies and how accessible it is. Only then would surface missions—rovers or eventually astronauts—arrive at carefully chosen drilling sites, reducing risk and improving the odds of success. This is not about replacing existing spacecraft. It is about adding a tool that makes the entire system more precise and efficient.
The stakes are substantial. Water ice on Mars is not merely a scientific curiosity. It is a record of the planet's climate history, potentially holding evidence of past life. For future human missions, it is a resource: drinking water, oxygen production, even agriculture. Targeting the right locations could unlock both scientific discovery and practical survival. The drone approach builds on lessons learned from NASA's Ingenuity helicopter, which proved that powered flight is possible in Mars' thin atmosphere and opened the door to more capable aerial platforms.
Aguilar framed the work as filling a gap between what we can see from orbit today and what astronauts will eventually observe on the ground. By adapting techniques developed for studying glaciers on Earth, scientists are making the hunt for buried ice on Mars something far more practical and precise. The next Mars mission that needs to find water will have better tools to do it.
Notable Quotes
If you want to make decisions about where to drill on Mars, you need to know if the ice you're trying to find is under one meter of debris or 10. That's the kind of information a drone-based system could provide.— Roberto Aguilar, lead researcher, University of Arizona Lunar and Planetary Laboratory
We are filling the gap between today's orbital observations and a more distant future, where astronauts land on Mars and make observations on the ground.— Roberto Aguilar
The Hearth Conversation Another angle on the story
Why does it matter so much whether ice is one meter down versus ten meters down? Can't a rover just drill deeper?
A rover carries limited power and equipment. Drilling through ten meters of hardened rock and debris is exponentially harder than drilling through one meter of loose material. You might run out of energy, or the drill might break. If you know beforehand that ice is shallow at a particular site, you can actually reach it.
So the drones are doing reconnaissance—scouting ahead for the rovers.
Exactly. Orbiters have been telling us where ice exists for years. But it's like knowing a city has water somewhere underground without knowing if it's in the basement or fifty stories down. Drones let you narrow that down to a specific block, a specific depth.
Why test this on Earth glaciers first? Why not just send drones to Mars?
Because you need to validate the method works before you spend millions sending hardware to another planet. The glaciers in Alaska and Wyoming are debris-covered ice, just like what we think exists on Mars. Testing there lets you prove the radar actually detects what you think it detects, and it lets you figure out the operational details—how fast to fly, what altitude, what direction.
What happens if this works? Does it change how we explore Mars?
It changes the strategy. Instead of orbiters finding ice and rovers drilling blindly, you have three layers: orbiters identify regions, drones map them precisely, then rovers drill at the best spots. It's more efficient, less wasteful, and it improves your chances of finding what you're looking for—or finding signs of past life in the ice itself.
And this is all because of a small helicopter that flew on Mars a few years ago.
Ingenuity proved the concept. It showed that flight works in thin air. Now we're asking: what else can we do from the air? What else can we learn? The drone with radar is the next step in that conversation.