What Happens If The Snowpack Fails Two Years In A Row
In 2021 the Upper Colorado River Basin got about 80 percent of its normal snowpack, which sounds like a mildly disappointing winter. Streamflow that spring came in at roughly 30 percent of average. That gap between what fell and what arrived is the single most important thing to understand about consecutive dry years in the American West, because it is not a straight line. The first bad winter is absorbed by reservoirs and by the ground. The second one is not, and the reason is that the ground itself has become a competitor for the water.
Here is what actually happens, in order, when the snow fails twice.
The Snowpack Is A Reservoir
Not a metaphor. Colorado's snowpack stores roughly 15 million acre-feet of water at its spring peak, which is a volume comparable to a major reservoir, and it releases that water on a schedule nobody has to manage.
On snowmelt-dominated rivers in the West, something like 60 to 80 percent of annual streamflow originates as meltwater, and 70 to 80 percent of the annual runoff arrives in the four months from April through July. In Colorado, 83 percent of water use depends on surface water fed by that melt.
So the snowpack is not a nice-to-have on top of the water supply. For most of the interior West, it is the water supply.
Year One Is Survivable
A single bad winter is what the entire storage system exists to absorb.
Reservoirs carry water across years specifically so a dry season does not become a shortage. The Colorado system was built to hold roughly four times its average annual flow for exactly this reason. Farmers pump a little more groundwater, cities lean on carryover storage, and the year passes.
What year one also does, quietly, is dry out the soil.
Year Two Is Where It Breaks
The second failure is not twice as bad as the first. It is worse than twice, and the mechanism is soil moisture.
When snow melts, the water does not run straight into a stream. It has to pass through the ground first, and dry soil absorbs an enormous amount before anything reaches a channel. After a dry year, the soil column enters winter with a deficit, and the following spring's meltwater goes into filling that deficit rather than into the river.
Colorado's assistant state climatologist Becky Bolinger has put the consequence plainly: in years that start with a water deficit, melting snowpack saturates the soil first, and to get a normal runoff season into the reservoirs you need an above-average snowpack.
Read that again. After a dry year, average snow is no longer enough to produce average water.
The 2021 Numbers
The clearest demonstration on record, and it followed a dry 2020 with no meaningful monsoon and extreme heat.
Upper Basin snowpack came in around 80 percent of normal. Runoff came in around 30 percent. A 20 percent shortfall in snow produced a 70 percent shortfall in water.
For comparison, hydrologists describe an efficient year as one where the two roughly track: 50 percent snowpack yields about 50 percent runoff. In 2021 the conversion rate collapsed, and the reason was the state of the ground the snow was sitting on.
The Feedback Loop
Dry soil does not just absorb water. It changes the weather above it.
Bolinger described it as a loop of hot soils driving evaporation and hot, dry air. Moist ground uses incoming solar energy to evaporate water, which cools the surface. Dry ground has no water to spend, so the energy goes into heating instead. Hotter air dries the soil further, and the cycle reinforces itself.
That is why a second dry year tends to be hotter than the first, and why the heat then accelerates everything below.
The Melt Comes Too Early
Warmer conditions move the melt forward in the calendar, and timing matters almost as much as volume.
In 2025 Colorado's statewide melt ran about 10 days ahead of the median, with sites in the San Juans and Sangre de Cristos melting out 20 days or more early. By early May, 24 percent of the state's SNOTEL sites had already melted out entirely.
Early melt delivers water before demand peaks and before reservoirs are positioned to capture it, and it lengthens the dry season on the far end. A given amount of snow arriving in April is worth less than the same amount arriving in June.
Less Snow Means More Heat
There is a second feedback loop running on the surface, and it is about reflectivity.
Snow reflects sunlight back to space. When the snow-covered area shrinks, particularly across the low and mid elevations between roughly 4,000 and 7,000 feet, more of that energy is absorbed by bare ground instead. That raises local temperatures, which melts more snow.
Satellite records for the Colorado basin show snow-covered area at its lowest in years, concentrated at exactly those elevations. Less snow makes it warmer, which makes less snow.
Groundwater Does Not Cover The Gap
The obvious backstop turns out to be a second casualty rather than a solution.
Research in the East River basin in western Colorado has found that mountain aquifers themselves fall in a warming climate, and that they can take streams down with them. Those aquifers feed baseflow into headwater streams year-round, so when they drop, the rivers lose water even outside the melt season.
The study's central finding is the relevant one here: warming threatens the ability of those aquifers to recharge after a single extremely dry year. Recovery is not automatic.
Reservoirs Absorb It, Then Show It
Storage hides the first year and reveals the second.
In 2021 Blue Mesa, Colorado's largest reservoir, hit a record low after water was sent downstream to keep Lake Powell generating hydropower. That is the visible mechanism of a second dry year: upstream reservoirs get drained to protect a downstream one.
In 2022 Reclamation went further and delayed a scheduled release from Powell to the Lower Basin, the first time the agency had held back water owed downstream to Arizona, California and Nevada.
Hydropower Goes Before Water Supply
The first thing a two-year failure threatens is not drinking water. It is electricity.
A dam generates power only while the reservoir stays above its penstock intakes, and that elevation sits well above the point at which water can no longer be released at all. So the sequence in a prolonged shortfall runs: reduced generation, then no generation, then delivery problems.
That is why the emergency measures in 2021 and 2022 were framed around protecting power production at Glen Canyon. Losing the turbines happens first and costs the Upper Basin states revenue that funds the entire system.
Agriculture Absorbs The Cuts
When deliveries do get reduced, farms take the loss before cities do, and this is by design rather than accident.
Under prior appropriation, rights are ranked by seniority, and much of the West's most senior water is agricultural. But cutting a farm is politically and practically simpler than cutting a city, so shortage responses lean on fallowing, deficit irrigation and voluntary compensated reductions.
Roughly 70 to 80 percent of Colorado River water goes to agriculture, which means the only place to find a meaningful volume quickly is fields.
One Good Year Does Not Fix It
This is the part that surprises people, and 2023 proved it.
That winter Colorado's statewide snowpack topped 140 percent of average, with the San Juan and Dolores basins near 179 percent. It was a genuinely exceptional year after a historically bad run.
Reclamation still forecast Lake Powell finishing the year at just 32 percent of capacity. A single outstanding winter refilled soil moisture and stabilized the system. It did not refill the reservoirs, because the deficit accumulated over two decades is far larger than one year of surplus.
Where It Stands Right Now
As of March 2026, snow water equivalent across the Upper Colorado Basin was running at 62 percent of normal for the date, with snow-covered area at its lowest in years.
One useful nuance from that same assessment: autumn rains in the southern part of the basin left soils wetter than in recent years, which means a higher proportion of the meager snowpack should actually reach the streams. The conversion rate matters as much as the total, and this year it may work in the basin's favor.
That is a thin consolation attached to a 62 percent snowpack.
The Sequence, In Order
Put the whole thing together and a two-year failure unfolds along a predictable path.
Winter one delivers below-normal snow. Reservoirs cover the gap and soil moisture is drawn down. Summer runs hot and the monsoon underperforms, deepening the soil deficit. Winter two delivers below-normal snow onto dry ground, and the runoff conversion collapses so that even the snow that fell does not arrive. Upstream reservoirs are drained to protect downstream ones. Hydropower generation is threatened. Delivery reductions are declared, and agriculture absorbs most of them.
None of that requires a catastrophic winter. It requires two mediocre ones in sequence.
The Honest Answer
What happens if the snowpack fails two years in a row is that the West discovers its water supply is not linear.
Eighty percent snow can produce thirty percent water. Average snow after a dry year is a shortfall. And the buffer that makes the first failure survivable is the same buffer that gets consumed making it survivable, which is why the second one lands on a system with nothing left underneath it.
The uncomfortable version is that the region has already run this experiment. It ran it in 2020 and 2021, and the response was emergency releases from upstream reservoirs, a delayed delivery to three states, and a set of interstate negotiations that are still unresolved.
