Seasonal outlook
Western snowpack outlook
How much water will be in the western snowpack on March 1 and April 1? An outlook of snow water equivalent as a percent of the 1991–2020 normal, with the odds of a low, near-normal or high snowpack year, for the Colorado-Utah mountains, the northern Rockies, the Sierra and Cascades, and the Southwest.
Issued
Each January, from December data. First live issue January 10, 2027.
Early read
A November issue for the northern Rockies and the Southwest only.
Not forecast
Snowfall totals. This is the water in the snow on the ground, not inches of snowfall.
Why this works when snowfall forecasts do not
By late December much of the season’s snow is already on the ground, and the rest of the winter’s precipitation leans with the Pacific. Together they give a usable read on the water that will be in the snowpack at the start of the melt season, months before the melt.
What an issue looks like
This is what the outlook would have said in January 2025, from data through December 2024, beside what the NRCS SNOTEL stations then measured. It was a middling year for the outlook, shown as it came out: close in the northern Rockies and the Sierra-Cascades, too high in Colorado-Utah, and right about a dry Southwest but not how dry.
| Region | March 1 snowpack | April 1 snowpack | What SNOTEL measured |
|---|---|---|---|
| Colorado-Utah | 110% 10% above normal 303238 | 107% 7% above normal 303733 | 77% Mar 1 78% Apr 1 |
| Northern Rockies (MT ID WY) | 104% near normal 313633 | 112% 12% above normal 283735 | 98% Mar 1 98% Apr 1 |
| Sierra-Cascades (CA NV OR WA) | 94% 6% below normal 243342 | 93% 7% below normal 233641 | 99% Mar 1 107% Apr 1 |
| Southwest (NM AZ) | 62% 38% below normal 323533 | no April snowpack | 14% Mar 1 11% Apr 1 |
| Upper Midwest (MI WI MN) | 121% 21% above normal 224236 | – | for information |
| Northeast (NY VT NH ME) | 71% 29% below normal 304030 | 90% 10% below normal 283933 | for information |
Bars show the odds that the region’s snowpack ends in the low, middle or high third of the previous 30 years; the largest share is darker. The expected percent is the central estimate; the odds are the better guide to risk. The two can point different ways because they use different yardsticks: the percent compares the expected snowpack with the 1991–2020 average, while the odds compare it with the last 30 winters, which in much of the West have run below that average and include a few very big years. So a snowpack a little below the long-term average can still rank as a typical or even high recent winter, and in the Southwest a year near 60 percent of the average is close to a normal recent winter. Regions are weighted by each climate division’s normal snowpack, so the high mountains count most. The Upper Midwest and Northeast are shown for information.
Track record
Tested over 1982–2025 and checked against 844 SNOTEL stations: how closely the January outlook tracked the March 1 snowpack each year (correlation, where 1 is perfect and 0 is no relationship), and how often it called the direction right.
| Region | 1982–2011 | 2012–2025 | Direction right |
|---|---|---|---|
| Colorado-Utah | +0.30 | +0.59 | 65% |
| Northern Rockies (MT ID WY) | +0.68 | +0.26 | 82% |
| Sierra-Cascades (CA NV OR WA) | +0.30 | +0.49 | 76% |
| Southwest (NM AZ) | +0.62 | +0.07 | 53% |
Skill score against climatology across the snowpack map: +13.9 for 1982–2011 and +10.8 for 2012–2025 (0 is no better than climatology). The northern Rockies were the strongest region in 1982–2011 and have been weaker since; Colorado-Utah and the Sierra-Cascades have improved since 2012. The Southwest swings the most from year to year and has been the weakest lately.
← The three-month temperature and precipitation outlook · How it verifies
The Amphibian snowpack outlook is an independent product. It is not affiliated with or endorsed by NOAA, NRCS or the National Weather Service. Snowpack history: NRCS SNOTEL and ERA5. Contains modified Copernicus Climate Change Service information [2026]; neither the European Commission nor ECMWF is responsible for any use that may be made of the Copernicus information or data it contains.