• New AI tool offers nationwide water quality predictions.
• Potential benefits include improved drinking water safety.
• Technology applicable across the Colorado River Basin states.

Tuesday, March 18, 2025 — A groundbreaking artificial intelligence (AI) tool developed by researchers at the University of Vermont (UVM) promises to revolutionize water quality forecasting across the United States. This technology, detailed in a study published on March 4, 2025, in the Journal of the American Water Resources Association, marks the first-ever use of the National Water Model specifically to predict water quality conditions.

Andrew Schroth, the lead researcher from UVM, highlights the significance of this innovation: “This new tool can be implemented across the country and broadly utilized by folks that could use water quality forecasts in any number of applications.” Schroth emphasizes that the adaptation of the National Water Model for water quality forecasting has “opened a new window that can really benefit the country as a whole moving forward.”

Testing the Tool in Real-World Conditions.

The team tested this advanced forecasting method on New York City’s extensive drinking water supply system, specifically focusing on the Esopus Creek catchment in the Catskill Mountains. Esopus Creek flows into the Ashokan Reservoir, which is a critical water source providing approximately 40% of New York City’s daily drinking water supply.

The primary concern in this region is turbidity—a measure of water cloudiness caused by sediments. “When too much sediment comes into the reservoir during or after big storms, New York City has to limit supply and modify their operations,” Schroth explains. The AI-driven forecasts offer a critical advantage by predicting turbidity levels, which enables water supply operators to respond effectively to storm-related disruptions.

National Implications, Including the Colorado River Basin.

Although initially tested in New York, the technology holds vast potential for application throughout the U.S., including the states dependent on the Colorado River Basin. This new model provides critical insights for managing water quality concerns such as sediment, nutrient runoffs, and algae blooms, all prevalent issues across the southwestern states.

The research, supported by the Cooperative Institute for Research to Operations in Hydrology (CIROH)—a collaboration involving the University of Alabama, NOAA, and the U.S. Geological Survey—integrates extensive datasets from the National Water Model with real-time stream sensors to predict water quality accurately.

Dr. John Kemper, now at Utah State University, underscores the broader impact, stating: “Turning a streamflow forecasting tool into a water quality forecasting tool paves the way for increasingly available forecasts to serve community needs.”

Practical Applications Across Industries.

This predictive capacity has numerous practical applications. Drinking water providers can better manage operations during storms, agricultural sectors can anticipate water availability, and communities can proactively respond to health threats from algae blooms or contamination events.

With the potential to significantly enhance water management practices nationwide, this AI-based forecasting system may become a vital tool for safeguarding water resources and public health in the Colorado River Basin states and beyond.

Journal Reference:

1. John T. Kemper, Kristen L. Underwood, Scott D. Hamshaw, Dany Davis, Jason Siemion, James B. Shanley, Andrew W. Schroth. Leveraging High-Frequency Sensor Data and U.S. National Water Model Output to Forecast Turbidity in a Drinking Water Supply Basin. JAWRA, 2025 DOI: 10.1111/1752-1688.70011

~~~

Image:  

 AI Artificial Intelligence concept, by Jernej Furman from Slovenia.  Licenced under the Creative Commons Attribution 2.0 Generic license.

Related posts:

Agencies to secure Rio Grande water amid El Vado Dam repairs BLM to boost monitoring with a $17.6 million investment Arizona seeks teams to bring new water supplies