Predicting the unpredictable
In a hurricane, timely, accurate information saves lives. Forecasts that pinpoint when a storm will strike, how strong it will be and where it is headed give people the time to act.
In the 1960s and 1970s, researchers at the NSF National Center for Atmospheric Research (NSF NCAR) helped develop dropsondes, small instruments that can be dropped from aircraft into storms to collect real-time data on temperature, pressure, humidity and wind speed.
Dropsondes soon became a critical forecasting tool. By 1982, the National Hurricane Center began using them during hurricane reconnaissance missions, leading to a 20 to 30% improvement in predicting a hurricane's path. Over the following decades, NSF NCAR continued to refine the technology, making dropsondes more accurate, reliable and cost-effective.
In the 1990s, NSF-funded researchers also helped develop Doppler on Wheels — truck-mounted mobile radar systems that collect high-resolution data from inside hurricanes and tornadoes. These systems have been used during major storms, including Hurricane Ike (2008) and Hurricane Helene (2024), to better understand wind dynamics, rain bands and storm intensification.
Credit: C. Semmer/UCAR
Engineering stronger defenses
NSF is also helping communities build stronger infrastructure that can withstand nature's worst, such as:
Harnessing the wind
The NSF Natural Hazards Engineering Research Infrastructure (NSF NHERI) Wall of Wind Experimental Facility at Florida International University uses its powerful fans to simulate Category 4 hurricane conditions, testing how winds up to 157 miles per hour affect residential and commercial buildings.
Their discoveries have improved roofing systems, influenced building codes and developed technology that protects structures and generates power during storms.
The power of waves
The Oregon State University O.H. Hinsdale Wave Research Laboratory, supported by the NSF NHERI program, aids researchers in studying storm surges and waves — everything from the physics of fluid dynamics to what happens when a tsunami hits a bridge.
By simulating hurricane-driven waves on scale models of coastal communities, they can predict real-world structural damage, informing the design of stronger buildings, bridges and wave barriers to reduce storm impact before it reaches shore.
Speeding up recovery
In the wake of hurricanes and other natural disasters, NSF has supported critical research to improve emergency response, recovery and long-term resilience.
Improving response after Hurricane Andrew
Following Hurricane Andrew in 1992, NSF-supported researchers assessed the storm's damage and evaluated Florida's emergency management systems, leading to significant improvements in building practices, disaster readiness and response strategies.
Their work also fostered stronger coordination among local, state, federal and private sector partners to enhance their response capabilities.
Better levee designs after Hurricane Katrina
In the aftermath of Hurricane Katrina in 2005, researchers supported by the NSF Small Grants for Exploratory Research program were on the ground within days to assist with emergency response, collect time-sensitive data and study the storm's immediate impacts.
Their work was instrumental in identifying the engineering failures that caused widespread flooding, leading to critical improvements in levee design and flood protection systems and helping reshape how communities prepare for and recover from future disasters.
Assessing damage from Hurricane Helene
After Hurricane Helene in 2024, NSF Rapid Response Research teams deployed to Florida's Big Bend region to collect ground and aerial data using drones and lidar, helping officials assess storm surge, flooding and infrastructure damage, providing crucial insights for communities to rebuild and prepare for future storms.
At the same time, two NSF-supported Extreme Events Reconnaissance/Research (EER) teams were mobilized to perform damage assessments and collect time-sensitive data on building performance and failures.
Supporting emergency response during Hurricane Beryl
After Hurricane Beryl struck Texas in 2024, Resilitix Intelligence, an NSF-supported startup, used artificial intelligence and digital imaging to assist emergency responders in affected communities — providing critical real-time information about food distribution centers, hospital access and neighborhoods cut off from aid.
Looking ahead
NSF continues to invest in the next generation of breakthroughs to assist in weather prediction, resilient infrastructure and disaster recovery, ensuring communities are ready for the storms of tomorrow.
NSF NCAR is leading the way with innovations like the MicroPulse Differential Absorption Lidar, a cost-effective laser system designed to provide continuous measurements of water vapor, aerosols and temperature in the lower atmosphere. Various field campaigns incorporate this technology to significantly enhance real-time weather models and help emergency managers make life-saving decisions.
Supporting a national network
Through its Directorate for Geosciences (GEO), NSF supports cutting-edge radar systems, remote sensing suites and laboratory facilities that span the nation.