WSU Researchers Develop Snow, Water Forecasting Tool

New AI-powered model could provide daily and weekly forecasts of water availability from mountain snowpack

Jan. 27, 2026 at 8:31pm

Got story updates? Submit your updates here. ›

Researchers at Washington State University have developed a new forecasting tool that uses artificial intelligence to predict daily and weekly snow-water equivalent, a critical metric for water management decisions in the Western U.S. The model outperforms existing approaches for about 90% of locations for daily forecasts and 70-80% of locations for weekly forecasts.

Why it matters

Snow-water equivalent is a key indicator of potential water availability, as 50-80% of annual streamflow in the Western U.S. originates from melting winter snowpack. Accurate forecasts can help water managers make better short-term decisions about flooding and long-term plans for irrigation, hydropower, and fisheries.

The details

The researchers used AI to create a forecast model that incorporates both temporal and spatial data from over 500 snow measurement sites across the Western U.S. Unlike current approaches that rely on historical data, this model can fine-tune predictions to relevant future scenarios. The model also provides valuable information on forecast uncertainty to help water managers make more informed decisions.

  • The researchers recently presented their work at the Association for the Advancement of Artificial Intelligence Conference in Singapore.

The players

Krishu Thapa

First author on the work and a graduate student in WSU's School of Electrical Engineering and Computer Science.

Kirti Rajagopalan

Assistant professor in the School of Biological Systems and a co-author on the paper.

Bhupinderjeet Singh

Co-author who completed his doctorate on this work at WSU.

Ananth Kalyanaraman

Director of the School of Electrical Engineering and Computer Science and co-author.

Washington State University

The university where the researchers developed the new forecasting tool.

Got photos? Submit your photos here. ›

What they’re saying

“Snow-water equivalent is critical for decision making because it tells you how much water would be available from the melted snow, which would go through streamflow or watersheds.”

— Krishu Thapa, Graduate student

“What this forecasting does is take that to the next level. Instead of just looking at years in the past, we can fine-tune our model into a smaller subset of future states that are relevant.”

— Kirti Rajagopalan, Assistant professor

“The most important thing is how confident we are about those predictions because the decisions that water managers make are going to impact people.”

— Bhupinderjeet Singh, Doctoral graduate

“I think that having an integrated way of being able to forecast interdependent variables is important. The current technology is to mostly predict variables individually. If we can tie that up in a much more integrated fashion, that would represent an advancement in the current way in which this information is being used.”

— Ananth Kalyanaraman, Director

What’s next

The researchers are working to build a dashboard to provide real-time forecasting that water managers can use. They also want to integrate weather forecasting and streamflow forecasts into their model.

The takeaway

This new AI-powered forecasting tool represents an advancement in the ability to predict water availability from mountain snowpack, which is critical for water management decisions in the Western U.S. By incorporating both spatial and temporal data, and providing information on forecast uncertainty, the model can help water managers make more informed and impactful decisions.