Q&A: Defining the “snow deluge” and projecting its future
Scientist Adrienne Marshall explains why these extreme snow years are expected to decline in our warming world.
I’m excited to share this story that I recently wrote for The Water Desk, a program at the University of Colorado Boulder’s Center for Environmental Journalism that I co-direct.
For California’s Sierra Nevada, the winter of 2022-2023 delivered an epic snowpack that broke many records and busted a severe drought.
The exceptional season, dubbed the “snowpocalypse” by some, caused havoc during the winter and flooding later in the year while also replenishing reservoirs and making skiers happy—once the roads and resorts emerged from storm closures.
Both hazardous and helpful, the banner year was also of interest to snow scientists, such as Adrienne Marshall, an assistant professor of geology and geological engineering at the Colorado School of Mines in Golden.
Marshall was lead author of a paper published in April in the Proceedings of the National Academy of Sciences that introduces the term “snow deluge” to describe extreme snow years like the one California weathered.
I spoke with Marshall recently about the study and its conclusion that snow deluges are likely to decline in the decades ahead as human-caused climate change continues to warm the planet and make it more likely that rain, rather than snow, will fall.
The video below contains the full interview, and the Q&A that follows includes excerpts from the conversation, edited lightly for clarity and brevity.
Could you please summarize the paper and its main findings?
In this paper, we were interested in looking at what’s happening with our biggest snow years. The snow community has had a lot of focus in the last several years on snow drought: what’s happening with changes in our lowest snow years, and how much more frequent should we expect those low snow years to get? But when we focus on the low snow years, there’s a little bit of a risk that we neglect to look at what’s happening with our biggest snow years. So when we saw the 2023 water year come along in California, it provided sort of a great study to dive into this question of biggest snow years.
We argue that we should call these big snow years “snow deluges.” We looked at how rare the snow deluge in California water year 2022-2023 was. Essentially, we see that it was, statewide, about a 1-in-54-year event is our best estimate.
There was a lot of discussion at the time of whether it was a record-breaking water year or not. We saw that about 42% of sites had the highest April 1 snow water equivalent ever recorded. That was more record-breaking sites than we saw in any other water year in California.
We were also interested in the extent to which the 2023 snow deluge and others like it are caused by relatively cool versus wet conditions. And maybe not surprisingly, we saw that both cool and wet conditions are required to have these snow deluges. You can get maybe average temperatures and average precipitation, but you really couldn’t have a very warm year and get a snow deluge.
Finally, we wanted to know what happens to our snow deluges in warmer climate scenarios. So we looked at the outputs of some climate models and found that it looks like—on average across climate models—we should expect our snow deluge years to decline in terms of the amount of snow we see on the ground. But they do decline less on a percentage basis than our average years.
How did you come up with the definition for a snow deluge?
There’s a little bit of a parallel here to snow drought—and drought in general—where generally there’s a lot of academic debate over how we should define these things.
In this case, we used April 1 snow water equivalent because that’s the type of data for which we have the longest record. Snow surveyors have been going out into the mountains to observe the amount of snow at individual locations for as long as 100 years or so, but they could only do that historically on a few days of the water year and commonly did it April 1st. So that’s a bit of a decision based on data availability.
Then we used a definition of looking at a 1-in-20-year return interval—the event that you would expect to see once every 20 years, statistically on average—as our cutoff threshold for what we call a snow deluge.
Did any of these findings surprise you?
Probably the most surprising was that the 2023 water year in California—we looked over November to March—was anomalously cold. And that was a little unexpected, particularly in the context of the record-breaking warm global temperatures. As far as I know, that’s just sort of an anomaly with respect to the rest of the global trend.
In terms of how we should expect the snow deluges to change, there are two competing factors happening. One is as conditions get warmer, of course we get less precipitation falling as snow, more falling as rain, and so we expect our total snow accumulations to decline. The other sort of competing component is as we get warmer conditions, the atmosphere can hold more moisture, so there’s an increasing probability of more precipitation intensity. Those two factors might counterbalance each other to some extent. We didn’t know for sure what the net effect would be and saw that it looks like the net effect is the warming having a bigger effect in terms of our snow accumulations.
Is it likely we’ll ever see another deluge like last year’s?
The 1-in-54-year event statistic that we came up with means that each year—over that same spatial area (California)—there’s a little under a 2% chance of an event that big happening again. That said, given our findings with the climate models, it looks like as our emissions rise and warming continues over time, that percentage and that probability is decreasing. So it’s getting less and less likely that we would see a snowfall event that big again. So I’m inclined to say that, statistically, it’s not likely that we should expect to see another year that big.
What are you able to say about snow deluges in other parts of the West?
We did see a remarkable amount of consistency in terms of the climate model projections for changes in snow deluges. It was really across almost the entire Western U.S. that we saw the average of the nine climate models we looked at were projecting declines in snow deluges and smaller changes in the snow deluges than the median years. That was consistent across California, Colorado, and the rest of the Western U.S.
With atmospheric rivers, we hear they can be both beneficial and hazardous. Is it helpful to think of snow deluges in the same way?
It’s certainly true that they can be beneficial and hazardous, even if we were to just think about one industry, like winter recreation and skiing. Of course, it’s good to get lots of snow at your ski resorts. And there were parts of the year where people couldn’t access the ski resorts and they had to close.
Similarly, you might have snow deluges that are good at one time of year, like beneficial for our summer water supply, but detrimental to our infrastructure in the winter. Or, ecologically, it might be beneficial for one type of organism and detrimental for another. So they’re certainly complex in terms of their effects. I don’t know that you could isolate it to some quantity being good and then above that being bad because I think there’s a big dependence on good or bad for who, when and where, which would be great to dive into. It sounds like decades of research, too.
What are the implications for water managers?
A big snow year—a snow deluge—can certainly help refill our reservoirs, which we’ve seen. The fact that we see these (snow deluges) declining might suggest that we shouldn’t count on big snow years coming along periodically and saving us.
Snow tends to be more predictable from a water manager’s perspective in terms of when they’re going to get runoff because if we have precipitation falling as rain, then you’re just kind of constrained to the uncertainty of your seasonal weather, which is pretty hard to forecast. But if you have a whole bunch of snow sitting in the mountains above your watershed, you know it’s going to melt and the question is how much do you have out there, when will it melt, and should you release water now to maintain flood storage capacity or just expect that maybe you’ve already gotten most of your snow runoff for the year?
So losing that snow deluge is probably challenging from a water management perspective because those big snow years offer more predictability and more ability, potentially, to refill reservoirs than if you are just relying on rain, which is so much harder to predict on a seasonal scale.
Any other takeaways?
A very good question to think about is: how much can we save if we’re able to get ourselves on a lower warming scenario? So treat these projections of snow loss as very concerning projections to be aware of and that should inform our decisions about how hard we work to mitigate climate change, but not as fatalistic predictions. There is a lot that we can do societally to try to nudge ourselves ever towards those lower warming scenarios, and that is what we should be doing to save as much of our snowpack as we can.