A small uncrewed boat became the first to capture data from inside a Category 5 hurricane

On September 28, 2025, a four-foot wind-propelled boat called the C-Star became the first uncrewed surface vehicle to capture and transmit data from inside a Category 5 hurricane. It rode out 150 mph winds in Hurricane Humberto's eyewall and sent wind, pressure, sea surface temperature, air temperature, and humidity readings back to shore every two minutes (WLOX).

NOAA's Atlantic Oceanographic and Meteorological Laboratory confirmed the milestone, calling the C-Star "the first ocean robot to collect data in a Category 5 hurricane" (NOAA AOML). Two more C-Stars penetrated Humberto later that same day while the storm was still a Category 4.

Dropsondes have measured storm-core wind and pressure for decades, but a dropsonde falls through a storm once and is gone. The C-Star drifted through Humberto's eyewall on a wind-propelled hull, solar-powered sensors reporting every two minutes for as long as the boat stayed afloat. Continuous in-storm surface meteorology is a data type the National Hurricane Center has historically lacked in real time. The NHC's public advisory at 5 PM AST on September 28 confirmed Humberto at 145 mph and 937 mb in the open Atlantic (NHC public advisory 17).

Hurricane hunter aircraft can fly the storm, but they leave when they land. Moored buoys sit in fixed positions and frequently get dragged or destroyed by storm passage. The result is a sparse, episodic record of what is happening at the air-sea interface inside a hurricane, and a forecast community that has learned to work around the gap. Continuous in-storm surface meteorology from a platform that survives the eyewall would close part of that gap. A four-foot wind-propelled boat that does it on solar power, at a price that lets a university program operate seven of them, makes the gap closable on a research budget.

Oshen CEO Anahita Laverack told WLOX that the C-Star readings were referenced in a National Hurricane Center forecast discussion, the agency's narrative product that explains the reasoning behind each advisory (WLOX). A forecast discussion is the forecaster's running commentary, not the data pipeline. Operational hurricane forecast models assimilate dropsonde, satellite, and buoy data on fixed schedules, and the question of whether the C-Star's two-minute surface stream has been wired into that ingest pipeline is separate from the question of whether a forecaster quoted it in a paragraph. The C-Star's two-minute surface stream has not been confirmed as an input to those models, and the gap between a forecast-discussion reference and operational model ingest is the gap between a research milestone and a forecasting tool.

TechCrunch reported in January 2026 that the C-Star's Category 5 transit was a first for an ocean robot and that the program is now expanding (TechCrunch). The framing, first but as a beginning, matches what the data supports rather than the survival narrative the company name invites.

The C-Star is four feet long, wind-propelled, fitted with solar panels, and carries a sensor suite that reads air pressure, air temperature, relative humidity, sea surface temperature, and wind speed and direction. The hull is unanchored, drifting wherever the storm sends it; the solar array keeps the payload alive long enough to report. Oshen, a UK-headquartered firm founded around four years ago, designed the boat for persistent ocean monitoring rather than for one-off storm chases, and the design suggests a sensor platform meant to live in place for weeks at a time, not a hardened probe. Three units inside a single major hurricane on the same day is unusual even for a research program. NOAA's AOML page on the milestone lists seven C-Stars operating during the 2025 season under the University of Southern Mississippi's Gulf Blue Navigator Program, a partnership that places the boats across the Gulf of Mexico during storm threats. USM's role is the program operator; NOAA's is the forecast-data consumer; Oshen's is the platform builder.

The next phase, per Laverack, is a Gulf-wide fleet of C-Stars carrying acoustic and visual sensors, with data feeding the Coast Guard and forecasters. USM's Gulf Blue Navigator Program is now accepting applications for the next cohort (WLOX).

Continuous two-minute readings from inside storms approaching the Gulf coast would give forecasters a real-time look at the air-sea interface in the zone where storm intensity is set, a region where direct in-storm measurements have historically been one-off at best. The data will sharpen hurricane warnings before landfall only if the two-minute stream survives ingestion into operational models, the acoustic and visual sensors add anything dropsondes cannot, and funding survives past the demonstration phase. Coastal evacuation decisions turn on forecast track and intensity errors that have been stubbornly hard to drive down; in-storm surface data is one of the inputs the equation has been missing, and a constellation would matter only if the operational forecast community treats it as a data source rather than a press release.

A first-of-kind transit earns a first-of-kind headline. The program has to clear funding, sensor validation, and model integration before the second and tenth transits are guaranteed.