Harvey was an epic storm with unprecedented destruction. It dumped 33 trillion gallons of water on Texas and surrounding areas, and flooded an area the size of Lake Michigan, with some parts of Houston recording over 50 inches of rain. The storm displaced hundreds of thousands of people, and exposed many more to the health hazards of floodwaters contaminated with fecal matter and industrial toxins. Regretfully, approximately 60 individuals lost their lives.

This lower-than-expected death toll was not by chance, but partially reflects over a decade of improved preparedness efforts since Hurricane Katrina. But given the structural damages done to over 200,000 houses, businesses, and infrastructure, and the reports of health care and nursing home facilities requiring emergency evacuations of vulnerable populations, it’s evident that there are still enormous gaps in both city planning and preparedness and response. The good news, however, is that Texas can and will move forward, and technology has the potential to fill this gap—in Texas and the rest of the nation.

While it may be impractical to plan for a 1,000-year flooding event (a 1 in 1,000 probability of a flood that size in any given year), 500-year floods have almost become predictable events in Houston. Unfortunately, the city’s current (anti) zoning regulations and infrastructure are mismatched to the risk, and it requires updated storm water management, land use, and development policies. More broadly, the population living in coastal areas has increased 41% since 1963. The desire for proximity to water and coastal living seems to outweigh negative consequences. Those living in flood plains don’t seem to move to higher ground even after recovering from a flood, and new construction continues in known flood plains. Similarly, those cut off from regular medications, medical devices, and treatment programs demonstrate that the nation has not fully heeded the lessons for health care planning from the more recent Superstorm Sandy, which crippled New York and the surrounding area in 2012.

While Texans were stranded behind a wall of water or waiting to be rescued from their roofs and flown to safety, in addition to calling 911, they did what they and millions around the world do every day—communicated and reached out on social media. Rich or poor, old or young, they shouted for help in 140-character tweets and became portraits of the true despair and impact of the storm on Instagram and Facebook Live. But the question was, who was listening? Who would respond, and ultimately, who would save them before it was too late?

With the advent of big data, blockchain, and robotic artificial intelligence being debated on the national stage, it is time for the country to better apply available technology to disaster response. This topic was the subject of “Operation Dragonfire,” the code name for a significant public-private partnership convened by the National Voluntary Organizations Active in Disaster with startup support from the Centers for Disease Control and Prevention (CDC) after noting the failures in information sharing and resource mismatches during Superstorm Sandy. The partnership involved a vast array of disaster response entities, including government at all levels, the technology sector, socially conscious corporations, volunteer organizations, and disaster relief operations.

Inspired by the notion that dragon fire could never be extinguished once ignited, this collaborative partnership came together to design the blueprint for a real-time national system that could help make sense of millions of pieces of disconnected data, including social media, and transform that data into insights for quick and decisive action by those charged to protect us through a combination of machine learning, analytic and fusion methods, and human intelligence. The tool was to be accessible on portable devices like smartphones in order to collate and synthesize disaster information from multiple sources, link resources with those in need, and share validated information during emergencies. “Safety check” from Facebook (FB) started as a similar approach to leverage social media to drive insights during a disaster.

As seen in other disasters, social media is often the platform of choice during disasters. With Harvey, rescuers used social media to find victims who were stranded, while those who needed rescuing used social media to get out a message to rescuers. A highlight of the Harvey response was the ad-hoc use of Zello by the “Cajun Navy” to coordinate rescuers and monitoring of social media to relay messages by numerous volunteer organizations, such as Standby Task Force.

It is time for Homeland Security and Health and Human Services to fund and support a national robust platform informed by past experience, but enabled by modern technology to address the significant information challenges that are a major by-product of disasters. The platform must be able to gather information and align resources before the next epic storm. Time and information are critical factors in saving lives during any disaster. We have the technology capability to bridge this gap.

Our choice is clear because the risk of extreme weather events is real and impervious to denial: We must become better prepared as individuals, cities, states, and a nation for the increased risk that these extreme and costly weather events bring. We must build resilient communities so that we recover more quickly. We must implement mitigation and adaption policies to address climate change. Alternatively, we could build an ark for those of us who don’t populate Mars. But this time, we need to make it big enough to have room for robots, the Internet, and smartphones.

Ali Khan is dean of the College of Public Health at University of Nebraska Medical Center.