With help from artificial intelligence, West Virginia University researchers are speeding up environmental scientists’ access to time-sensitive data about ecosystem health, allowing experts to respond rapidly to events like droughts or wildfires.
Supported by an Early-Concept Grant for Exploratory Research from the National Science Foundation, Steve Kannenberg , assistant professor of biology in the WVU Eberly College of Arts and Sciences , is integrating AI directly into local monitoring sensors to help scientists track ecosystem health in near real time.
Every day, forests, grasslands, and ecosystems around the world “breathe,” constantly trading carbon, water, and energy with the atmosphere, Kannenberg explained.
These exchanges shape everything from local water supplies to how ecosystems respond to a warming climate. But Kannenberg added that it can take scientists months, sometimes years, for environmental data to become something they can use — a gap he and WVU postdoctoral researcher Jie Hu are working to close.
Scientists studying ecosystem fluxes rely on field structures known as eddy covariance towers, or flux towers. These are specialized weather stations equipped with ultra-fast sensors to collect data. But historically, researchers have faced a major bottleneck: By the time usable reports reach them, critical ecological events have long passed.
“We have amazing tools to measure how these ecosystems breathe, but because measurements are taken 20 times per second, processing this huge, tremendous amount of data into something that’s quality controlled takes immense effort,” Kannenberg said. “In these networks of flux towers, there can be a latency of months to years before data is ready to share with the community. It’s that issue this grant is trying to tackle.”
Currently, researchers must manually sift through massive datasets to quality control the results after specialized software runs initial complex algorithms. They must filter out sensor glitches caused by rain or wind, fill gaps in missing records, and tailor calculations to each specific site. These tedious processes slow the sharing of vital information.
“The traditional way to process this data relies on very established mathematical and physics-based equations that take a long time to calculate,” Kannenberg said. “We’re using AI tools to circumvent those complex equations and process the data much more quickly.”
As co-principal investigator, Hu is leading the development of a “processing pipeline” that rapidly turns environmental observations into ready-to-use data products.
“The main goal is to make these products easy to access by cutting the long delays in data release and providing interactive and visual diagnostics,” she said.
To enable researchers to watch ecosystems respond to events as they unfold, the project pairs AI with a national network of monitoring stations called the National Ecological Observatory Network, or NEON. Kannenberg also relies on “edge computing,” which uses very small, on-site computers to crunch numbers where and when they’re collected, instead of shipping data off to cloud servers.
“In the waiting period, any extreme environmental events can be missed. If there was a severe drought, we wouldn’t know how that’s impacting the ecosystem until a year down the line. If there was a wildfire, we wouldn’t know how much carbon was being burned off that ecosystem until years down the line,” Kannenberg said.
The instant data processing will give users a chance to catch transient “hot spots” and “hot moments” of biological activity, such as sudden bursts of plant growth after rain, localized drought stress, or nonbiological emissions from nearby vehicles, which are usually lost in traditional, delayed averages.
“Faster release of accessible flux data doesn’t just help scientists understand how ecosystems are responding to change. It also gives land managers and decision-makers a near-real-time view of environmental conditions,” Hu said.
Real-time processing also solves operational delays.
“NEON is a government-funded effort to have a standardized set of towers in major biomes across North America,” Kannenberg said. “A ton of money and time goes toward this, but they don’t know if there’s a problem with their instrumentation until someone physically collects the data and processes it.”
To remove the technical barriers, the team is building an interactive, plain-language AI chatbot interface. The tool allows anyone, from students to policymakers, to ask simple questions about the data, such as: “Was there a drought last year in West Virginia? How did that impact the forests?” The goal is to deliver scientifically grounded answers. Kannenberg also plans to integrate the interactive tool into his Ecosystem Ecology course at WVU.
They plan to test their approach at two very different sites: a grassland known for unpredictable bursts of activity at the Central Plains Experimental Range in Colorado and the Harvard Forest in Massachusetts, a temperate forest with a steadier seasonal rhythm.
“I’m excited to streamline the process and see how new technologies in artificial intelligence and edge computing can accelerate discoveries we haven’t even imagined yet,” Hu said.
Kannenberg said the value lies less in any single scientific discovery than in what faster access to data makes possible.
“When I first moved to Morgantown in 2023, there was a severe drought. Cheat Lake was visibly drying up, leaving boats sitting in mud,” Kannenberg said. “It’s difficult for scientists to study the impacts of sudden events like that because we must scramble to assemble teams and equipment after the fact. By having an automatic system that rapidly alerts us to ongoing environmental extremes, we can more easily have targeted field campaigns or measurement campaigns to better understand these events.”
The project builds on Kannenberg’s ongoing research into how forests and drylands store carbon and respond to a changing climate, including his recent work on the western United States’ 23-year megadrought .