On the morning of June 9th, Laura Lin was working from her home in Lanesville, a small town in southern Indiana near the Kentucky border. While she was on a Zoom call, heavy rain began to fall, eventually flooding her yard. Lin and her family escaped safely to a neighbor’s house, but the town received over 8 inches of rain in just a few hours—far above what is typically considered heavy rainfall. The water receded within a few hours, but the event highlighted the sudden and dangerous nature of flash floods. Floods are the second-deadliest weather-related hazard in the United States and the deadliest globally. Even 6 inches of fast-moving water can knock an adult off their feet, and 2 feet of water can move vehicles like cars and trucks. Climate change is increasing the frequency of extreme rainfall events, leading to more frequent and severe floods. In areas like Lanesville, where rainfall can quickly overwhelm local infrastructure, early warning systems are crucial. A new software called the Transient Artifact and Continuous Learning System (TACLS) is being developed to improve flood prediction. It uses satellite data and machine learning to detect areas at risk of flash flooding before they occur, helping the National Weather Service (NWS) issue more timely warnings. Laura Lin described how, during the recent storm, warnings came too late. "We didn’t get the 'get on your roof' warnings until I was already at that person’s house," she said. "All the people in town were already flooded when they started sending out alerts." TACLS has the potential to change how flash flood warnings are issued. Ivory Small, a science and operations officer at the NWS San Diego Weather Forecast Office, says the system could save lives. "Without TACLS, a storm moving into your area could kill some folks," he explained. "With TACLS, you can put out the warning and save some folks." Weather forecast offices across the U.S. are responsible for issuing alerts, and TACLS is designed to support them by providing real-time data on atmospheric conditions that could lead to flash flooding. Currently, the NWS uses a combination of rain gauges, satellite data, and weather radar to monitor flood conditions. Forecasters analyze this data along with historical information about soil types and terrain to determine if a flash flood is likely. The system has different levels of alerts, from flood watches—indicating a potential risk—to warnings, which signal a life-threatening situation. However, existing technology has limitations, such as less detailed satellite coverage over land and gaps in rain gauge networks in less populated areas. TACLS uses data from the Global Navigation Satellite System (GNSS), which is typically used for earthquake prediction, to measure moisture in the atmosphere. More moisture leads to longer delays in signals between satellites and ground sensors, providing clues about upcoming storms. Scientists from the University of California, San Diego, the NWS, and NASA developed the system, with funding from NASA’s Earth Science Technology Office. TACLS is currently being used in the LA and San Diego forecast offices and will soon be available nationwide. While TACLS is a powerful tool, it is not a complete replacement for human forecasters. It provides additional data to help them make more accurate decisions. Bhavik Chandna, a UCSD graduate student involved in the project, says the system is being trained using years of atmospheric data to recognize patterns in storms. Though it can sometimes produce false positives, the system is designed to work alongside human experts, not replace them. TACLS is expected to be especially useful in the Western U.S., where flash floods occur more frequently and quickly. However, Chandna believes the system could be adapted for use in other regions with the right combination of GNSS sensors and weather data. As research continues, the NWS aims to improve its ability to predict and respond to flash floods, potentially saving more lives in the future. For residents like Laura Lin, early warnings could mean the difference between safety and danger in a rapidly changing climate.