Scientists studying water supply focus on weeks following peak snowpack
April forecasts may no longer be reliable benchmark
Water managers in the Colorado River basin are gaining a better understanding that what happens in the weeks after peak snowpack — not just how much snow accumulated over the winter — can have an outsize influence on the year’s water supply.
Water year 2021 was historically bad, with an upper-basin snowpack that peaked around 90% of average but translated to only 36% of average runoff into Lake Powell, according to the U.S. Bureau of Reclamation. It was the second-worst runoff on record after 2002.
One of the culprits was exceptionally thirsty soils from 2020’s hot and dry summer and fall, which soaked up snowmelt before runoff made it to streams. But, those dry soils are only part of the story.
A new paper from the Desert Research Institute, a nonprofit science arm of the Nevada university system, found that heat waves in April 2021 drove record snowmelt rates at about 25% of snow-telemetry (SNOTEL) sites looked at across the West. SNOTEL is a network of remote sensing stations throughout the West’s mountainous watersheds that collect weather and snowpack information.
A heat wave that was concentrated over the Rocky Mountains on April 1-7 contributed to record snowmelt at 74 stations, including areas that feed the Colorado River.
Support Local Journalism
A few agencies release monthly water-supply forecasts for April through July, including the National Resource Conservation Service, the Bureau of Reclamation, and the Colorado Basin River Forecast Center. The April forecast is the first prediction of how streamflows will shape up for the year.
But, according to the paper, in 2021, “rates of snowmelt throughout April were alarming and quickly worsened summer-runoff outlooks, which underscores that 1 April may no longer be a reliable benchmark for western water supply.”
The paper did not quantify what exactly the record melt speed meant for water supply, but paper author and Associate Research Professor of Climatology Dan McEvoy said it definitely contributed to the poor inflow into the nation’s second-largest reservoir in 2021. It also shows there are many more factors relevant to predicting the water supply than just how much water is in the snowpack — a metric known as snow-water equivalent, which is measured by SNOTEL sites.
“There was a combination of things that was contributing to this really low runoff in places like the Colorado River basin,” he said.
Some of these other factors include very little April precipitation and warm nighttime temperatures, which didn’t allow the snowpack to get into the daily freeze-thaw cycle that’s common in the spring. Persistent high pressure kept skies clear and sunny, which meant that more of the snowpack sublimated, evaporating instead of turning into liquid.
“When it’s sunnier and warmer, you can lose some of that water directly to the atmosphere,” McEvoy said. “It doesn’t even get to melt and go into the runoff.”
These rapid melting events could also help set up prime conditions for wildfires, he said, something he wants to continue studying.
“When you have the snow disappear earlier, there’s more time with the ground exposed, which contributes to drying out the vegetation in the spring and summer and an earlier onset to wildfire season,” he said.
After peak snowpack
Climatologists at Colorado State University are working on a similar study that looks at how factors such as precipitation after peak snowpack affect spring runoff. Their findings underscore how important the conditions of the six to eight weeks after peak snowpack are for predicting streamflows.
“One of the things we found that was crystal clear from the study was that one of the major sources of water-supply forecast error is what happens after peak snowpack,” said Peter Bennett Goble, a climatologist at CSU who is working on the study. “Just knowing how much uncertainty is still out there on April 1 or even April 15 probably allows water managers to be a little more cautious, maybe hold a little bit more back, especially if it looks like it’s going to be an early runoff.”
Predicting whether reservoirs will fill — and, therefore, how much water to release to make room for the inflow — can be tricky. Some municipal water providers use the Colorado Airborne Snow Measurement Program — with its lidar-equipped planes — to more accurately measure snowpack. For example, Denver Water has used the measurement to see how much snow is in the headwaters of the Blue River basin, which feeds Dillon Reservoir, its largest storage bucket.
But, aside from this technology, which is expensive and not yet available everywhere, water managers rely heavily on data from the SNOTEL sites to make streamflow forecasts. This method has limitations, providing just a snapshot of conditions at one location.
These limitations can be seen in recent years’ forecasts for Ruedi Reservoir, on the Fryingpan River. Initial forecasts in April 2021 projected Ruedi could probably fill to its entire 102,373-acre-foot capacity, but the reservoir ended up only about 80% full that year. In 2020, each of the three main forecasting agencies also overpredicted Ruedi inflow for the months of April, May, and June. (An acre-foot covers 1 acre to a depth of 1 foot.)
Tim Miller, a hydrologist with the U.S. Bureau of Reclamation who manages operations at Ruedi, said his models predicted a 2021 Ruedi inflow of 111,000 acre-feet, but only 77,000 acre-feet actually flowed in. That the models are based on historical SNOTEL data from past decades is a drawback as climate change progresses, but it’s the best we have, Millheer said.
“It makes the assumption that what we have seen in the past is what we will see in the future, which is a really poor assumption when you’re in the middle of a change in the climate,” he said. “We will probably see events like we haven’t seen in the future, and we are using what we’ve seen to predict them.”
Dave Kanzer, director of science and interstate matters for the Glenwood Springs-based Colorado River Water Conservation District, said people often look for a single explanation when streamflows don’t match predictions. The River District both owns and operates Wolford Mountain Reservoir, near Kremmling, and as well as stores water in Ruedi Reservoir. But, there is often a whole host of compounding factors that water managers will have to begin weighing more heavily as the climate warms.
“It’s not just about soil moisture, it’s not just about solar radiation, it’s not just about temperatures, it’s not just about the winds — it’s everything,” he said. “In some cases, like 2021, you get what some people like to call the perfect storm.”