Innosphere spotlights NSF ASCEND-funded collaboration between Colorado State University and Virridy to deploy sensor networks and machine learning across Colorado watersheds.

Colorado's Cache la Poudre River spans 125 miles from Rocky Mountain snowmelt to the South Platte River, serving 400,000 people and 185,000 acres of agricultural land. Traditional water quality management has relied on infrequent grab samples, resulting in slow, reactive decisions based on incomplete data. Recent catastrophic wildfires have made the need for integrated warning systems more urgent than ever.

A New Approach to Watershed Monitoring

Associate Professor Matt Ross at Colorado State University has built a water-quality sensor network stretching from the Rockies to Interstate 25. Multi-parameter sondes measure pH, turbidity, and chlorophyll every 15 minutes, and the team is developing machine-learning models to forecast total organic carbon, chlorophyll a, and turbidity with 1–3 day lead times.

"The need for an integrated water quality warning system has become more urgent in the past five years as catastrophic wildfires have burned the water supply basins for large parts of the western United States," said Ross.

The Lume Sensor

Professor Evan Thomas and the Virridy team developed the Lume sensor, which uses specific wavelengths of light to detect contaminants like E. coli through tryptophan-like fluorescence. The Lume can detect previously hard-to-measure contaminants at a fraction of traditional sensor costs, and has been deployed on Boulder Creek and the Yampa River, scaling from 20 pilot units to 200 commercial units.

Collaboration Through NSF ASCEND

"One of the benefits of having the support of the NSF ASCEND Engine is having the ability to come collaborate with Evan and spend a lot of time thinking deeply about how our projects align, how we can help each other, how our teams can be integrated," said Ross.

Approximately one-third of Colorado's wastewater treatment utilities discharge into rivers not meeting Clean Water Act standards. The integrated sensor and AI/ML approach could enable green infrastructure solutions to replace built infrastructure along waterways.

Looking Ahead

Ross aims to create similar monitoring systems for other western partners facing increasing water stress, enabling proactive rather than reactive watershed management across Colorado and Wyoming.

Read the full feature at Innosphere.