top of page
Unexpected-Texas_Port-Lavaca_Pier.avif

Data-Driven Flood Intelligence for Texas Communities

In partnership with Texas Sea Grant and Texas A&M University-Corpus Christi, Calhoun County is using Hohonu’s real-time sensors to turn local knowledge into actionable flood intelligence - an adoptable template to expand statewide

When Beryl came through, we looked at the sensor reading and saw the water receding. We knew then the storm wasn’t coming. Having that data even before the national alerts in real-time gave us a sigh of relief — we could tell people to shelter in place.​

​

- Calhoun County Judge, recounting his experience in July 2024

The Story

Calhoun County needed a better way to see floods as they happen, from unexpected sunny-day road coverage to high tides and storms. Traditional gauges missed localized inundation, leaving responders with blind spots during high tides and storms.

​

With support from Texas Sea Grant and Texas A&M University - Corpus Christi Conrad Blucher Institute, Hohonu’s real-time sensors now feed both emergency operations and TAMU-CC's AI-driven flood forecasting. TAMU-CC is using Hohonu’s continuous feed to train AI models capable of forecasting flood conditions hours in advance - supporting both local decision-making and state-level resilience planning.

5

Locations

Providing hyperlocal flood visibility where none existed before

24/7 Monitoring

Always-on water level data streamed to county officials and researchers

3 Institutions

County, university, and private sector united in real-time resilience

image.png

Track of Hurricane Beryl, June 28 – July 9 2024.


Beryl made landfall on the Texas coast as a Category 1 hurricane, bringing coastal flooding and heavy rainfall to Calhoun County before moving inland toward the Midwest.

Screenshot 2025-10-16 at 8.10.05 PM.png

At the height of Hurricane Beryl, Calhoun County officials watched the Six Mile Park sensor show waters falling - which signaled that the community could shelter-in-place rather than evacuate

Problem

  • Flat coastal terrain makes even small changes in water level and rainfall flood local roads and neighborhoods.

  • Existing NOAA, TCOON, and USGS gauges are too distant to capture neighborhood-scale flooding.

  • Emergency managers lacked timely, localized data to inform road closures and evacuations.

  • Rural communities receive limited or delayed coverage compared to major coastal cities.

  • Researchers needed continuous, real-world measurements to validate and improve AI flood models.

Stakeholders

  • Calhoun County: uses real-time water level data to guide local response and road closure decisions.

  • Texas Sea Grant & TAMU-CC: integrate sensor data into flood modeling and AI-based forecasting research.

  • Hohonu: provides hardware, data infrastructure, and API integration for continuous water level visibility.

Outcomes

  • Real-Time Decision Support: County emergency managers accessed live water level data during Hurricane Beryl to guide local alerts and response.

  • AI Model Training: Continuous sensor data now fuels TAMU-CC's flood forecasting and machine learning research.

  • Improved Local Forecasting: On-the-ground data validated and refined predictive models for surge and inundation.

  • Cross-Sector Collaboration: County, university, and private partners created a shared data framework for future coastal resilience efforts.

Solution

  • Real-Time Data Feed: Hohonu sensors stream continuous water level data to county dashboards and research servers at TAMU-CC.

  • AI-Ready Dataset: High-frequency readings standardized and archived for machine learning applications in flood forecasting.

  • Collaborative Network: County, Sea Grant, and university teams coordinate data access and validation across multiple coastal sites.

  • Expansion Planned: Additional sensors planned for nearby Gulf counties to strengthen Texas’ early warning and resilience capabilities.

bottom of page