Climate Change and Variability in Snowpack

The American West is expected to see less mountain snow and that snow will melt earlier as our climate warms under human-caused climate change. Multiple studies have come to this conclusion. However, very few studies have looked at how this disappearing snow is likely to change from year to year.

Addressing this problem is a recent study published in the journal Geophysical Research Letters. The study, led by University of Idaho researcher Adrienne Marshall and including CIRC’s very own John Abatzoglou, examines not only how the amount of snow the American West can expect to see under climate change is likely to dwindle, but also the interannual variability—the year-to-year change—this dwindling amount of snow is likely to undergo.

Under projected future climate change, the study concludes, the West is likely to see less snow overall and more back-to-back low snow years. When the snow does show up, the peak amount of that snow is likely to come earlier in the calendar year than it has historically. And, the study notes, the amount and timing of that snow is likely to vary considerably from year to year under climate change…with one exception. Snow variability is likely to disappear entirely at many lower-elevation mountain sites as rising temperatures force precipitation to fall as rain rather than snow. All these findings spell trouble for water management here in the American West.

In the West, mountain snow, or snowpack, acts as a natural water reservoir. By slowly melting over the summer months and flowing from the mountains downstream, snowpack provides water for farmers, fish, and city dwellers alike, and it does this during our region’s normally warm summer months. In the Northwest and California, these summers are not only warm, they also tend to be pretty dry. In other words, we really rely on water from melting snowpack to get us through those dry months.

Marshall and colleagues’ results depict a future climate based on what is likely to happen if we continue emitting greenhouse gases at our current rate. This emissions scenario is called RCP 8.5.

As our frequent readers know, RCP 8.5—sometimes called the “business and usual” greenhouse gas emissions scenario—helps climate researchers simulate a world in which nothing (or virtually nothing) is done to cut greenhouse gases and, as a consequence, global warming continues throughout this century and beyond. As far as emissions scenarios are concerned this is literally the worse case scenario.

Marshall and colleagues were principally interested in how the high warming expected under RCP 8.5 would impact a measure called snow water equivalent (SWE).

SWE is essentially a measure of how much water is held in snow. SWE is frequently used by water resource managers to determine how much water can reasonably be relied on from a given amount of mountain snowpack. In much of the West generally and in the Northwest in particular, having an adequate amount of SWE during the winter and spring months often translates into having adequate streamflow for wildlife and irrigation during the summer months. In other words, getting good SWE numbers is pretty darn important.

How climate change—and RCP 8.5 in particular—is likely to impact SWE has been the subject of multiple previous studies. However, while other studies have looked at the American West’s disappearing mountain snow—see “Snow Declines in the American West”—Marshall and colleagues is the first study to examine future changes in both the magnitude and timing of mountain snow variability from year to year. By examining the projected future effects of RCP 8.5 on SWE at sites located on mountains across the West for the future period 2050-2079, the researchers were able to parse this out.

The researchers sliced up their SWE data by looking at the maximum amount of SWE a site was projected to see each winter under RCP 8.5 as well as the timing of that maximum amount of SWE. As you might image, a warming climate means less snow and the high warming expected under RCP 8.5 means considerably less snow. And that is true. However, some of the details of this change are worth paying attention to.

Examining their maximum SWE numbers, the researchers found that extremely low snowpack years—years sometimes called snow droughts—are likely to increase substantially. Snow droughts are periods when regions receive what, compared to historical numbers, are very low amounts of snow, for example, like much of the Northwest experienced in 2015.

Snow droughts can occur because above-normal temperatures force precipitation to fall as rain instead of as snow, or because not enough precipitation has fallen to create an adequate amount of snow, or through a combination of both warm temperatures and low precipitation levels. All these varying types and combinations of snow droughts are likely to contribute to the back-to-back low snow years we see under RCP 8.5, according to Marshall and colleagues.

But that’s just the big picture. The details of this change are also noteworthy.

The researchers found that the occurrence of consecutive years with snow droughts, that historically (1971-2000) occurred just 6.6 % of the time, were modeled under RCP 8.5 to occur 42.2% of the time by the middle decades of this century.

The timing of that maximum amount of SWE is a slightly different but no less discomforting story. By the middle decades of this century, the timing of maximum SWE—that is the day that the peak amount of SWE occurs at a mountain site over a given winter—is expected to come earlier in the year, but that’s not all.

The researchers also found that the timing of maximum SWE varied significantly from year to year at least for some mountain ranges. These changes, note the researchers, were greatest in Sierra Nevada and Cascade ranges. However, this pattern didn’t hold over all.

Marshall and colleagues found that year-to-year variably of SWE over the whole American West is actually likely to go down, not increase. The reason for this is simple enough. As temperatures increase under the high warming scenario, many lower elevation mountains that historically received their water as snow will increasingly see that water come as rain. In other words, while in the past precipitation might have come as rain one year and then snow the next, future warming means that that variation is likely to disappear and rain will dominate. Hence variability goes down. (You can’t, after all, get less than zero).

Marshall and colleagues end their paper with what has become a familiar refrain in papers like this one: a call for water resource managers to start preparing for what, at least as far as emissions scenarios is concerned, is the worse case scenario.

Publication: Marshall, Adrienne M., John T. Abatzoglou, Timothy E. Link, and Christopher J. Tennant. “Projected changes in interannual variability of peak snowpack amount and timing in the Western United States.” Geophysical Research Letters(2019). https://doi.org/10.1029/2019GL083770.

Feature Image: Photo Caption: SR 20 – North Cascades Highway – Washington Pass, Dec. 1, 2008. (Photo Credit: Washington State Department of Transportation, some rights reserved.)

Nathan Gilles is the managing editor of The Climate Circulator, and oversees CIRC’s social media accounts and website. When he’s not writing for CIRC, Nathan works as a freelance science writer. Other posts by this Author.