Advancing Flood Monitoring and Forecasting in Texas

Listen here on Spotify | Listen here on Apple Podcast

Episode released on September 10, 2026 
Episode recorded on April 30, 2026

David Maidment talks about flooding challenges in Texas and the U.S. and approaches to reduce flood hazards. 

David Maidment is a Professor Emeritus at the Cockrell School of Engineering, and he served as the Director for the Center for Water in the Environment for 14 years. His research focuses on surface water hydrology with a particular emphasis on applications of GIS to hydrology.

Highlights | Transcript

David Maidment’s presentations in chronological order with ppt slides are available at this link. 

1. Kerrville July Fourth Flooding in Texas

1. Kerrville flood highlighted a preventable tragedy 
2. David Maidment described the July 4th Kerrville flood as one of the deadliest disasters in Texas history, resulting in the death of 137 people. He mentioned that this was the largest loss of life experienced in Texas after the New London natural gas explosion in a school in 1937 that resulted in 295 deaths of teachers and children. He emphasized that improved communication and use of existing flood information could have reduced loss of life. He identified a critical "last-mile problem" in getting scientific information to those who need it most. He published an OpEd in the Dallas Morning News on August 12th titled “Texas is a leader in flood research, but more coordination is needed.”
3. David Maidment testified before the Texas Legislature on July 31, 2025, highlighting that accurate river data is essential for saving lives. His testimony is available online (link). 
4. He emphasized the need for a unified flood intelligence system. Although forecasting, monitoring, mapping, and modeling capabilities have improved dramatically, they remain fragmented. David Maidment advocates for a coordinated flood intelligence framework linking federal, state, county, and local systems to support real-time decision making. 
5. Real-time flood information should reach drivers directly: he envisions flood warnings integrated into navigation systems such as Waze or Google Maps based on flood map nowcast allowing vehicles to be rerouted away from flooded roads and low-water crossings, similar to how traffic congestion is handled today. This would provide actionable information. 
6. Personal tragedy motivated his flood research: the death of Travis County Sheriff's Deputy Jessica Hollis during a flood in 2014 profoundly influenced David Maidment's work. Her death reinforced the importance of providing actionable flood information to emergency responders and motorists.

2. Flood Monitoring Programs in Texas

1. Road elevation mapping (REM) provides a critical missing layer: UT researchers developed a statewide road elevation model using high-resolution lidar data, adding elevation ("z") information to Texas roads and bridges—over 470,000 miles of roads. This allows identification of vulnerable low-water crossings and flood-prone transportation corridors.  Texas has 25,000 span bridges over water. Maps are based on lidar data from Texas Geographic Information Office (85,000 points/mile, processed into 1 x 1 m grids). This product allows flood scenario modeling by overlaying flood modeling layers with the Road Elevation Models to determine which bridges and roads will be flooded during severe events. (Fig. 1). The REM showed that the flooding in Loop 410 at Perrin Beitel area in San Antonio appears as a dip in the road, increasing flood risk. This flood resulted in 11 people losing their lives. 
2. Pin2Flood is a mobile app that David Maidment’s group developed to allow on scene first responders to develop real-time flood inundation maps. The application uses a hydrology calculation called HAND (Height Above Nearest Drainage). This measures elevation contours relative to the stream bed. By pairing a dropped pin with pre-calculated HAND models, the app instantly estimates water coverage at one-foot depth increments. (Zheng et al., JAWRA, 2018)
3. Bridge monitoring networks provide early warning: Working with Texas Dept. of Transportation (TxDOT), David Maidment helped develop the FAST (Flood Assessment System for TxDOT) network using radar sensors installed on bridges that measure both water level and flow velocity (link). The network includes 80 gauges, the largest network of radar gauges in the U.S. (Fig. 2). 
   • During the Kerrville flood, velocity measurements downstream of Kerrville, provided velocity signals nearly an hour before peak stage arrived (Fig. 3). 
4. Texas Tech's expanded weather network will improve forecasting: Texas Tech Univ. and its National Wind Institute received  $24 M from the legislature (Senate Bill 5) to expand the West Texas Mesonet system into Central Texas and the Hill Country and improve atmospheric modeling. The expansion will include 40 new 33 ft mesonet stations and automated rain gauges. This system is called the Texas Weather Prediction and Measurement System. Three high-resolution radars will also be installed to monitor storms (Fig. 4)

3.  State and National Flood Programs

1. Texas Convection Permitting Ensemble Forecast System: Texas Tech Univ. (Brian Ancell) is developing regional ensemble forecasts (100 members) at 2 km grid spacing with 15 min output frequency. Brian Ancell is also developing TX Warn-On-Forecast System (WoFS) with 50 ensemble members for high impact events with up to 6 hr forecasts at 1 km grid spacing and 5 min output frequency (Ancell, FIRO presentation, 2026). Need to convert large ensembles of rainfall forecasts to 100 flood map scenarios—"real-time flood engineering." The National Weather Service has a Warn-On-Forecast system that did not deploy during the Hill Country flooding but now Texas will have its own system.
2. Simple models often outperform complex models operationally: He emphasized that national-scale operational forecasting requires computationally efficient approaches, such as rating curves and HAND methods. These simplified methods make real-time flood mapping feasible across hundreds of thousands of river miles. 
3. National Water Model transformed U.S. hydrology: David Maidment helped launch development of the National Water Model in 2014–2015, creating the first operational system capable of forecasting streamflow nationwide in near real time. This represented a paradigm shift comparable to the evolution of modern weather forecasting.
4. National real-time flood inundation mapping (FIM) is nearing completion: by September 2026, the National Weather Service will complete real-time FIM across the entire United States, providing a major national capability for forecasting flood extent and impacts (Fig. 5)  (link).
5. Flood inundation maps correspond to: 
   •National Water Model latest streamflow analysis
   •Maximum stage in the 5-day forecast from the official, traditional NWS streamflow predictions
   •Maximum stage in the five-day forecast guidance from the National Water Model
   •Static categorical flood inundation mapping (i.e., CatFIM) for select NWS river forecast locations within the flood inundation mapping domain
6. Tradeoffs in flood forecasting: David Maidment argues that future flood forecasting must balance three objectives: accuracy, reliability, and speed. Traditional engineering models often prioritize accuracy but are too slow for emergency response applications. 
7. Data assimilation is essential for reliable forecasts: Future systems must continuously incorporate real-time observations from stream gauges, rainfall measurements, and soil moisture networks to correct forecasts as conditions evolve, improving forecast reliability during rapidly changing flood events.

4.  Need for Integration

1. Hurricane Harvey demonstrated the need for integrated information: While working in the Texas State Operations Center during Harvey, David Maidment observed both a lack of information early in the disaster and an overwhelming amount of disconnected information later. This experience reinforced the need for a single, comprehensive operational flood map. 
2. Future progress requires stronger local partnerships: Following the Kerrville disaster, David Maidment has worked closely with Schreiner University, local communities, emergency managers, and first responders. He believes successful flood resilience depends on combining improved science, local knowledge, training, and effective communication to translate research into operational action. 
    

Overall takeaway: The podcast highlights how Texas has become a national leader in flood forecasting and mapping, but the greatest challenge now is translating sophisticated scientific capabilities into timely, actionable information that saves lives during rapidly evolving flood events.

The book “Geowater: A Geographic Approach to Water Data and Forecasting” will be published in September 2026 (Fig. 6).  “Geowater is a call to action. It challenges readers to rethink how we use data and technology to confront the growing challenges of water scarcity, flooding, and environmental change. By linking global processes to local consequences, Maidment shows that a geographic approach to water is key to building a more resilient and sustainable future.”