In the Western U.S., snowpack is an important tool for predicting annual streamflow from April through July. But Ben Livneh, a physical hydrologist and professor at the University of Colorado, says it's time to look for new ways to predict streamflow.
"There's a lot of risk associated with not anticipating a drought," Livneh said during the Oregon-Washington Water Year 2020 meeting Oct. 29.
His presentation was one of many weather-related topics at the annual meeting, which covered highlights from the 2020 water year that ended on Oct. 1, and forecasts for the 2021 anticipated La Niña winter ahead.
Snowpack is used extensively throughout the West to predict streamflow, which is one of several factors used to predict drought, Livneh said. But snowpack is already declining across the West due to climate change.
To find out whether measuring snowpack will become less useful, he and a colleague looked at 28 climate model projections, and used a hydrology model to simulate streamflow, snowpack and other data currently used in snow-dominated areas. They discovered that with continued reliance on snowpack to predict drought, by midcentury, seasonal drought will be less accurately predicted for 69 percent of areas dominated by snowmelt. That will increase to 83 percent by late century.
Their findings were published in the study, "Drought less predictable under declining future snowpack," published in the journal Nature Climate Change in April.
Livneh said areas most likely to have inaccurate predictions are in low elevations. Those above 10,000 feet continue to have accurately predicted streamflow and drought using snowpack. He said positioning snowpack data-collection sites at higher elevations is one way to help predictive accuracy.
Adding in other predictive data, such as precipitation, temperature and soil moisture, appears to improve drought predictions, he said. By including other data, "only a third of areas see decreased predictive skill, down from 83 percent," he said, adding, "That's very heartening, and a suggestion that forecasts should be more inclusive" of other data in snow-dominant areas.
Livneh is now working on a project to determine what other data can help predict drought in the future.
Mark Brusberg, chief meteorologist for the U.S. Department of Agriculture, gave opening remarks at the meeting, and talked about some impacts the Pacific Northwest is already experiencing from climate change.
He said that the Northwest is 1.5 degrees Fahrenheit warmer compared to the last century, and that temperatures could increase between 3 and 10 degrees over the next century.
April 1 snowpack is down over the last 40 to 70 years; the cool season over the Cascades is 2.5 F warmer, and peak runoff is sometimes a month earlier compared to 50 years ago, he said.
Kirby Cook, meteorologist for the National Weather Service in Seattle, gave an overview of this winter's forecasts. La Niña has been in place since August, and the latest outlook brings an 85 percent chance that La Niña will continue through winter, and a 60 percent chance it will persist through the spring, Cook said.
He said most models are forecasting a moderate La Niña, and a few predict it will be strong.
La Niña is quite often associated with cooler and wetter than average winters across the Pacific Northwest. "It doesn't say how cold, or how much wetter. It's very broad expectations over the whole season," he said.
For November, the outlook favors above normal precipitation over much of Washington, and equal chances of normal, above-normal or below-normal precipitation in Oregon and Northern California.
California temperatures are expected to be above normal, with equal chances of normal, above-normal or below-normal temperatures across most of Washington and Oregon, he said.
For the three-month period from December through February, forecasters are calling for below-normal precipitation across southern states and above-normal precipitation across the northern states. The greatest chances of above-normal precipitation are in Washington and the northern half of Oregon, with the highest probabilities in northeastern Washington, the Idaho panhandle and most of Montana.
Overall temperatures during those three months are expected to be below normal throughout Washington.
Cook said the outlook doesn't project seasonal snowfall accumulations, and this winter's snowpack will largely depend on the storm track. "Generally, snow is not predictable beyond seven days, and in the lowland, 36 hours," he said. Still, the outlook is favorable for snow, he added.
Larry O'Neill, Oregon state climatologist, said that at the beginning of the 2020 water year, on Oct. 1, 2019, the region's water supply was in very good shape. But then drought conditions expanded across the Pacific Northwest in the fall, and with some recovery in January, drought again expanded throughout the summer.
By the end of the summer, he said, only 6 percent of Oregon was free of drought, and about one-third of the state was in severe drought. Fifteen counties received state-level drought declarations.
O'Neill said Oregon ended the year with an average of more than 24.4 inches of rain, although rainfall was not distributed equally.
Washington saw more abundant water conditions, with just 6 percent of the state showing severe drought.
Looking at precipitation over the past 125 years, the 2020 water year ending on Sept. 30 was the 15th driest, and California experienced its 13th driest. For other Northwest states, the 2020 water year ranked 22nd driest in Montana, 27th driest in Idaho and 50th driest in Washington, he said.
Oregon also experienced its 10th warmest year on record, while Washington had its 19th warmest over the 125-year period, O'Neill said