Centralizing national flood data in the cloud

Researchers from the University of Texas collaborated with other researchers, federal agencies, commercial partners, and first responders to create the National Flood Interoperability Experiment (NFIE). They used Microsoft Azure to help build a prototype for a national flood data-modeling and mapping system with the potential to provide life- and cost-saving information to the public. The goals of the NFIE include standardizing data, demonstrating a scalable solution, and helping to close the gap between national flood forecasting and local emergency response.

Sharing flood information for better prediction and response
In October 2013, the Onion Creek area near Austin, Texas, faced a particularly destructive flood. While onsite studying the flood, David Maidment, professor of civil engineering at the University of Texas, spoke with Harry Evans, chief of staff for the Austin Fire Department. They realized that they had similar goals for improving flood prediction and response, and could collaborate well with their different areas of expertise.

Maidment, who specializes in hydrology and flooding at the Center for Research and Water Resources, brought together participants from academia, federal agencies, commercial partners, and first responders to create the National Flood Interoperability Experiment (NFIE). He wanted a technology infrastructure that would allow flood information to flow in from various agencies and academia, and then flow out to allow citizens and first responders to better understand what was happening.

“What we're trying to do in the National Flood Interoperability Experiment is to prototype a set of infrastructure and services that can communicate with one another and with the public in a uniform and open way,” says Maidment.

Microsoft Azure for data analysis, storage, and sharing in the cloud
Microsoft Research helped the NFIE find the computational power it needed in Microsoft Azure. The NFIE uses Azure to perform statistical analysis of present and past flood data to help design a prototype for a national flood data-modeling and mapping system with the potential to provide life- and cost-saving information to the public.

By using Azure, the NFIE can standardize and store data in the cloud. Maidment and colleagues at the University of Texas developed a new language that provides both a common way to store time-value pairs, like river flow time, and a standard way of communicating that information through the Internet. The US Geological Survey adopted this language to publish its time-series data on water observations, and the National Weather Service will also use it to publish forecasts as part of the NFIE. When this common language is implemented operationally, those organizations will be able to communicate and collaborate more efficiently with one another.

More flood information provides protential for improved public safety
NFIE uses Azure to deliver more forecasts than any one agency could. Currently, the National Weather Service makes forecasts at about 3,600 locations on rivers in the country. The NFIE expects to demonstrate delivery of specific and actionable data for 2.67 million locations nationally, including smaller streams. It also expects to increase the spatial density of flood forecast locations by a factor of more than 700, compared to the current National Weather Service system.

Ultimately, the NFIE has demonstrated that information with a greater level of detail has the potential to increase real-time responsiveness that can improve public safety and save lives. Working closely with the Austin Fire Department, the NFIE shows how data can be used to improve decision-making. Evans notes that this work will help the NFIE develop a template that agencies can use nationwide, along with their threat and risk analyses, to help communities better protect themselves from the risks of flooding.