Major dust-up for water in the Colorado River

Nearly all water in the 1,450-mile-long river originates from snowmelt in the Upper Colorado River Basin, a region adjacent to arid landscapes that regularly blow dust onto mountain snowpacks. The darker snow absorbs more sunlight, speeds up melting and depletes snowpacks earlier, leaving downstream communities at risk of running dry.

Currently no snowmelt models account for dust -- that's a big problem for water managers who rely on accurate estimates to responsibly allocate water flowing through their districts.

A University of Utah-led study debuts a powerful remote-sensing dataset that informs the timing and magnitude of snow darkening and the impacts on melt rates across the Colorado Basin, in real time. The research is the first to capture how dust effects the broad expanse of headwaters feeding the Colorado River system.

The authors analyzed 23 years of daily satellite images to observe patterns of snow darkened by dust deposition during the melt season during April and May from 2001-2023. The findings revealed that dust-driven melting tends to peak earliest and be most intense in the central-southern Rocky Mountains at mid-alpine elevations.

Across the board, dust accelerated snowmelt every spring, even in less dusty years. Spring melt rates, measured at observation sites, are typically around 10-15 mm per day. The new study found that dust deposition can speed snowmelt by up to 1 mm water-equivalent per hour when the sun is at its peak. In high-dust years, that adds up to10 mm of melt per day that can be directly attributed to the darkening effect.

"It's not just how much dust gets deposited over a season, but also the timing of dust deposition that matters," said Patrick Naple, doctoral candidate of geography at the U and lead author of the study. "Dust is very effective at speeding up melt because it's most frequently deposited in the spring, when days are getting longer and the sun more intense. Even an extra millimeter per hour can make the snowpack disappear several weeks earlier than without dust deposition."

As the most comprehensive evaluation of dust-driven snowmelt to date, the study's insights could help produce a more accurate understanding of dust on snow impacts across the full basin, as well as improve forecasting and water allocation for a system under extreme pressure from changing climate and populations.

"The degree of darkening caused by dust has been related to water forecasting errors. The water comes earlier than expected, and this can have real world impacts -- for example if the ground is still frozen it's too early for farmers to use. A reservoir manager can store early snowmelt, but they need the information to plan for that," said McKenzie Skiles, associate professor at the U's School of Environment, Society and the study's co-lead author. "If we can start to build dust into the snowmelt forecast models, it will make water management decision-making more informed."

The study was published in Geophysical Research Letters on March 9, 2025.

Eye in the sky

Spring storms kick up rich, red dust from across the Colorado Plateau and carry it for miles until hitting mountain slopes and dropping deposits onto the snowpack. The dust darkens the normally bright surface, leading to faster melt.

Previous research has recorded dust-driven melt at individual study sites, but none have measured an area as large as the Colorado River's headwaters that cross multiple states.

To level up, the researchers expanded on a concept hatched by Skiles and co-authors Thomas Painter and Annie Burgess in 2012 that utilized satellites to remotely measure snow darkening. They used the Moderate Resolution Imaging Spectrometer (MODIS) instrument on the Terra satellite, which records multiple wavelengths of solar radiation reflected from landscapes below its orbit, every day, to build detailed maps of land-surface characteristics.

The researchers focused on the amount of the sun's energy reflected by a surface, known as albedo. The brighter the surface, the higher the albedo, and the more energy it reflects. The darker the surface, the lower the albedo, and the more energy it absorbs. Across all land cover types, snow naturally has the highest albedo.

The authors developed algorithms that did two things: quantified how dust lowers the albedo for every single pixel of satellite imagery; and calculated how the additional energy absorbed by dust impacted melt rates.

What drives the dust?

The data threw some curveballs. The latter half of the study's 20-year period, 2014-2021, had slightly less dust-driven melting than did the first half, 2001-2013.

"We were pretty surprised because the western U.S. has had an ongoing drought that's made the region increasingly arid over the last few decades," said Naple. "Intuitively, you'd think that if the soil is drier, there's going to be more dust emissions, but that's not what we saw."

Consequential dust-on-snow events depend on multiple factors coalescing in just the right way; wind speed, soil moisture, vegetation, surface disturbance and precipitation timing will determine how much dust is picked up and transported. Though not conclusive, the authors think that there may be more vegetation and lower wind speeds in dust-source regions based on work by colleagues. The factors contributing to lower dust impacts could be related to climate variability, but the 20-year time scale is too short to capture climate-level patterns.

"We can use remote sensing data to get a real time estimate of how much any given dust event is impacting the snow cover, but it doesn't give us much lead time because we don't know in advance when we're going to get a big dust event," said Skiles. "If we're better able to understand the drivers of dust emission, we could predict how that would impact the snow for the rest of the season."

More research is needed to link land-use change, climate scale variability and other factors that influence dust emissions and transport.

"We know from sediment core records that dust deposition in this area skyrocketed following modern settlement of the West. That tells us that the level of dust we see today is directly related to human activity," said Skiles. "If we could track ongoing land-use changes and surface disturbance in the region, we might have a better shot of predicting large dust events."