For more than a decade, researchers at the National Oceanic and Atmospheric Administration (NOAA) have pursued better ways to forecast two of the biggest concerns for boaters: hurricanes and rip currents. Recent advances in drone technology and coastal modeling are finally allowing those long-standing research goals to become practical tools for mariners, coastal managers, and the public.
“The technology is challenging, and anytime you’re promoting something new in the government, it’s challenging,” says Joe Cione of NOAA’s Atlantic Oceanographic and Meteorological Laboratory Hurricane Research Division. “But I feel very good about this.”
Cione leads NOAA’s effort to use unmanned aircraft to study hurricanes. Since the mid-2000s he has worked to find ways to collect detailed measurements from parts of storms that manned aircraft cannot safely reach. Early tests used the heavier Aerosonde drone, which weighed around 25 to 30 pounds. Cione’s team—working with partners from the U.S. Navy and NASA—launched those vehicles from the ground and waited for them to return with data. In several attempts they succeeded only a few times, including flights into Hurricane Ophelia in 2005 and Hurricane Noel in 2007.
Ophelia was the first time a drone entered a tropical cyclone, Cione notes, but the operation illustrated the limitations: long flights with only a short time actually inside the storm. That low efficiency drove the team to consider different launch methods and lighter platforms.
Engineers borrowed an idea from a Navy technique used to deploy probes. Instead of launching drones from the ground, they experimented with deploying them from aircraft already flying over storms. In 2009 the team tested a 13-pound Coyote drone, shooting it from a plane and getting it to fly. The breakthrough still required further development, but it established a viable path forward.
A major turning point came in 2012 after Hurricane Sandy prompted congressional support for expanded hurricane research. Funding allowed NOAA to accelerate drone development and testing. In 2014 the team successfully air-deployed a Coyote into Hurricane Edouard, proving the concept under extreme conditions: a small aircraft surviving deployment from a fast-moving plane and operating in very strong winds.
By 2018 Cione concluded NOAA needed purpose-built platforms that could fly farther and remain inside storms for longer periods. The agency solicited designs from small businesses and encouraged multiple approaches, a development strategy similar to how diverse teams pursued different Covid-19 vaccine solutions. The goal: multiple reliable drone types that NOAA can deploy as mission needs dictate.
One of those new systems, the Altius drone, completed a test flight over Maryland in January. Cione says the Altius met nearly all performance objectives, earning what he describes as a B-plus. The remaining work focuses on optimizing reliability and endurance before routine deployment into hurricanes.
When fully operational, these drones will deliver a dramatic improvement in storm sampling. Manned hurricane reconnaissance aircraft typically fly at about 10,000 feet and release instrument dropsondes—brief snapshots that collect vertical profiles of the atmosphere as they descend. In contrast, an Altius or comparable drone can maneuver repeatedly through different parts of a storm, gathering continuous measurements across the eyewall and inner core. That continuous record is like replacing a single photograph with a motion picture, offering far more context for forecasting and emergency decision-making.
“It’s a snapshot versus a movie,” Cione says. That richer data stream should sharpen forecasts used to make evacuation decisions, issue marina and harbor advisories, and inform boat owners whether their vessels are likely to be safe in place or should be moved well before a storm arrives.
While drones address offshore and storm-core measurement gaps, NOAA scientists have also advanced rip-current prediction closer to shore. Senior Scientist Greg Dusek began studying rip-current risk about a decade ago with two major hurdles: collecting reliable data in surf zones, where waves break and sand shifts constantly, and using high-resolution nearshore models that could turn measurements into forecasts.
Dusek found a practical data source in local lifeguard records. The Kill Devil Hills Lifesaving Station in North Carolina had long kept detailed accounts of rescues and hazardous rip-current conditions. Those observations provided a valuable baseline for when rip currents become dangerous to swimmers.
Combining lifeguard data with ocean observations, Dusek’s team developed an initial rip-current model in 2013. At the time, running that model in real time wasn’t possible because operational nearshore forecasting systems were still under development. As the National Weather Service completed the Nearshore Wave Prediction System—a suite of high-resolution wave and water-level models covering U.S. coasts—the pieces came together.
Tying the rip-current model to the Nearshore Wave Prediction System allowed NOAA to validate forecasts in locations such as Miami and San Diego and to partner with other lifeguard agencies to refine performance. In February NOAA moved the model into operational use, giving forecasters a way to predict the likelihood of hazardous rip currents on a 1 to 100 percent scale—similar to probability-of-precipitation forecasts used in routine weather apps.
Boaters and beachgoers will begin to see rip-current risk appear in weather products and local forecasts as NOAA distributes the data. The agency is also exploring enhancements such as incorporating webcam feeds from waterfront hotels and surf shacks and applying artificial intelligence to video imagery to identify rip-current signatures in real time. Those advances could further improve short-term warnings and beach safety messaging over the coming years.
Together, improved storm-sampling drones and operational rip-current forecasts represent meaningful steps forward in coastal safety and marine forecasting. For boaters, coastal communities, and emergency managers, these tools promise clearer, more actionable information—helping people make better decisions and reducing risk when hazardous weather and ocean conditions approach.