Drought has become increasingly difficult to predict in a warming world, as snowpack—which typically provides early warning for drought—diminishes. So NOAA is funding a new CU Boulder-led project that will develop new techniques for drought prediction that do not rely purely on snow-based methods, harnessing alternative techniques to improve scientists’ ability to predict and respond to drought. 

“In our changing climate, snow is projected to no longer hold the predictive power it has today, but yet we still heavily rely on snow information to predict drought,” said Ben Livneh, lead investigator on the new project, CIRES Fellow and Assistant Professor of Civil, Environmental and Architectural Engineering at CU Boulder. “We need to develop new techniques to advance our ability to predict drought in the future.”  

Normally, water from abundant melting snowpack flows predictably into mountain streams, and eventually on to lakes and reservoirs. But as Earth’s temperatures climb and snowpack dwindles, the relationship between snow and streamflow shifts. Sparse snowpack meltwater has to travel a farther distance before reaching streams, and an unpredictable volume of water can be lost due to evaporation. So Livneh and his team plan to explore other methods of drought prediction, including water models, soil moisture, precipitation and more.

The new project is funded with $500,000 from NOAA's National Integrated Drought Information System (NIDIS) through the MAPP program for three years of work. A key innovation will be the use of Machine Learning tools to find ways to improve current and future drought prediction. The team will collect direct input from regional stakeholders to help shape the modeling and Machine Learning work and assess the feasibility of alternative strategies. 

The findings will be presented at an interactive workshop to ensure findings are most relevant to decision makers. “We hope this work overcomes the challenges researchers have been faced with in a reduced-snow world,” said Livneh. “By innovating better drought prediction strategies we can better inform water management and planning for the western United States.”

The project’s co-investigators include Joseph Kasprzyk, also an Associate Professor of Civil, Environmental and Architectural Engineering at CU Boulder, and Benet Duncan, Managing Director of CIRES’ Western Water Assessment.