Researchers at the Technical University of Munich have flown a drone and laser system that maps the gas cloud over an active volcano with roughly 5 percent measurement error, and they ran the first fully autonomous deployment on Vulcano, an Aeolian island off the coast of Sicily. The target is the chemical signal that precedes an eruption, measured without sending a single volcanologist up the crater rim.
A Laser On A Cart Reads The Gas Cloud From Below
The setup splits the work between the ground and the air, funded under the German Research Foundation project Measurement Technology on Flying Platforms. A laser mounted on a ground cart automatically tracks a drone carrying a reflector, and the beam weakens as it crosses the volcanic gas between them. That fading signal is the measurement: the more gas in the path, the more the laser gets absorbed.
The drone, a 5.5 pound (2.5 kg) aircraft the team nicknamed Tina, flies a pre-programmed grid for 10 to 15 minutes at distances up to 197 feet (60 m), collecting as many as 3,000 measurements per flight. The platform can operate at altitudes up to 9,800 feet (3,000 m), which matters on volcanic terrain. An algorithm folds in local wind conditions and converts the absorption data into a tomographic map, a three-dimensional picture of where the gas sits and how concentrated it is at each altitude.
The approach solves a contamination problem that has bothered volcanologists for years. Gas readings taken at ground level do not come only from the volcano, because surrounding vegetation and soil emit carbon dioxide of their own. Lifting the measurement into the air isolates what the mountain itself is exhaling.
The CO2 To SO2 Ratio Rings The Warning Bell
The number that matters is a ratio. According to the TUM team, the relationship between carbon dioxide and sulfur dioxide in volcanic gas shifts in a recognizable pattern before activity: the ratio climbs sharply, then falls. Reading that curve in time gives authorities a window that seismic sensors alone do not provide.
The stakes are not academic on this particular island. In November 2021, Vulcano’s carbon dioxide output jumped from 80 tons to 480 tons a day, residents reported trouble breathing, and the mayor ordered the island’s roughly 250 permanent residents out of their homes at night. The island’s Grand Crater last erupted in the late 19th century, but the mountain never stopped exhaling.
TUM researcher Marius Schaab deployed the system on Vulcano, working alongside a team that includes Prof. Achim Lilienthal, deputy director of the TUM MIRMI robotics institute, volcanologist Nicole Bobrowski of Heidelberg University, and Prof. Thorsten Hoffmann of Johannes Gutenberg University Mainz, whose group works the complementary approach of mounting sensors directly on the aircraft. The underlying method was published in IEEE Sensors Letters in September 2025.
“Our goal is to automate the measurement and mapping processes and have artificial intelligence interpret the data,” Lilienthal said.
That sentence carries the roadmap. A monitoring station that flies itself, measures a whole gas cloud in 15 minutes, and flags the dangerous ratio shift without a human in the loop is the version of this technology a civil protection agency would actually buy.
Drone Volcanology Has Been Building To This
DroneXL covered gas sensor drones sampling volcanic plumes back in November 2020, when the state of the art was flying miniaturized spectrometers and sampling bottles through the cloud. The laser tomography approach maps the entire cross section instead of the single points an onboard sensor touches, which is the difference between a thermometer and an MRI.
The field work keeps proving the platform elsewhere. The USGS Hawaiian Volcano Observatory flew its uncrewed aircraft over Kilauea’s erupting summit crater last year, mapping lava and tephra from positions no geologist could stand. A DJI Matrice 600 with a thermal camera hovered directly over an active vent for close-up imaging that a helicopter crew would never attempt.
This is the job drones took first and still do best: going where a human body cannot afford to go. Fires, armed standoffs, even nuclear facilities where aircraft measure radiation levels. In this particular case the drone’s work is exactly that, reading dangerous numbers in a way that puts no human life on the line.
DroneXL’s Take
No sugarcoating this, an academic press release usually oversells, so let’s calibrate what Vulcano actually proved. One island, one campaign, a 5 percent error margin, and a method paper. Nobody has predicted an eruption with this rig yet, and the team does not claim otherwise. What they proved is the instrument: autonomous flights turning laser absorption into gas maps that used to require careers of risky fieldwork.
That is worth more than the headline suggests. The CO2 to SO2 ratio is established volcanology, and the bottleneck was always getting clean, repeated measurements without walking scientists into a crater. A system that does it every day, in any weather the drone tolerates, changes the data density that forecasting models eat.
And the pattern is bigger than volcanoes. We have watched drones fly into hurricanes for NOAA, chase data around severe storms, survey the seafloor, and search for survivors in earthquake rubble. The machines keep getting more useful and more adaptable, and the dangerous places keep getting less lonely.
Watch whether the team’s automation goal turns into a permanent monitoring installation on Vulcano or a sister volcano. The day one of these stations flags a ratio spike before an evacuation order, this stops being a research story.
Sources: TUM, Phys.org, Interesting Engineering
