Eyes in the sky: How NASA helps gauge drought impact

Shortly before 5 P.M. Pacific Daylight Time on Friday, a small, nondescript Beechcraft Air King climbed from the runway at Mammoth Yosemite Airport, reached 15,300 feet and headed over the snow-starved Sierra Nevada.

To anyone driving past the airport, nestled on the backside of the mountain range, the airplane's departure might have seemed like just another private plane taking off from a rural airfield.

Instead, the 1960s-vintage, twin-engine turboprop – NASA's Airborne Snow Observatory (ASO) – is playing a key role in helping water managers in drought-ravaged California track the amount of water stored in the state's paltry Sierra snow pack.

The aircraft carries two instruments whose data combine to provide the most comprehensive estimates yet of snow's water content – critical information for forecasting the mount of water that the mountains hold in reserve for what traditionally has been the state's dry season.   

The observatory began flying in 2013 as a three-year demonstration project, starting with one watershed. On Friday, the ASO would fly two sorties, traveling along tightly spaced, back-and-forth tracks over four watersheds.

The observatory's progress during its first two years has transformed from a let's-see-if-this-works effort to a must-have data source that has caught the attention of other Western states.

Until now, water managers have never known the true distribution of snow water equivalent across a watershed, says Thomas Painter, a hydrologist at NASA's Jet Propulsion Laboratory in Pasadena and the project's lead scientist.

"You can't manage what you don't measure," he says.

Yet mountain snows provide about 75 percent of the West's water. Population growth, a relentless draw-down of water stored in aquifers, and global warming's projected impact on precipitation and soil moisture pose significant challenges for managing water resources.

As if to underscore the point, researchers at NASA's Goddard Institute for Space Studies and Columbia University's Lamont-Doherty Earth Observatory published a study in February that yielded projections for "a remarkably drier future that falls outside the contemporary experience" of people and ecosystems in western North America.

California, entering the fourth year of a record-smashing drought for the state, has become the poster child for that future. A third of the state's water typically has come from the Sierra snows.

Across the Sierra Nevada, however, "the snow pack is about half of the previous lowest, driest on record,” says Frank Gehrke, who heads the California Water Resources Department's Cooperative Snow Survey Program. “We're down in uncharted territory." As of April 1, the average snow pack statewide is about 5 percent of normal for this time of year.

For the few watersheds the ASO has been measuring, data on the snow pack's water content are being fed directly into a new generation of steam-flow forecast models with "quite phenomenal improvement in water-supply forecasting," Mr. Gehrke says.

ASO data have reduced errors from between 20 and 40 percent to about 5 to 7 percent for forecasts looking 8 to 12 days ahead, Dr. Painter adds.  He and colleagues currently are evaluating the data's impact on projections of April-to-July run-off.

Even for watersheds with few ASO measurements and no upgraded forecast models yet, reservoir managers are eager to get the observatory's water-content data, Gehrke notes.

"Water managers are looking for any kind of information that could help them make reasonable decisions on reservoir management," he says.           

The reduction in errors is a big deal, notes Jeffrey Deems, a researcher at the National Snow and Ice Data Center in Boulder, Colo., and a member of the ASO team.

"Managing a dam for water supply and hydropower, you have to be careful you don't over-top things by not leaving enough room in your reservoir, but you don't want to leave so much room that you run out of water midsummer," he says.

Managing these conflicting goals in the face of large measurement errors – particularly in the face of a changing climate – is a major challenge.

ASO is providing far more precise knowledge of the amount of water that snow in each basin represents, which "enables this new paradigm where we can manage more efficiently and to a much tighter margin," he says.

Others have taken note. NASA and Colorado recently inked a Space Act Agreement to use the observatory for snow water-content measurements in the San Juan Mountains, notes Dr. Painter. The snow pack in the San Juans feeds the Rio Grande River and one of its key tributaries in Colorado, the Conejos River. The observatory was slated to fly east Sunday to begin making measurements there during the coming week.

Wyoming and New Mexico, along with the National Oceanic and Atmospheric Administration's Colorado Basin River Forecast Center, also have expressed an interest in the project, Dr. Deems says.

In addition, the ASO is scheduled this summer to build terrain maps for the entire Sierra Nevada in preparation for expanding snow water-content measurements to encompass the full extent of the range.

The observatory makes its measurements using lidar and an imaging spectrometer. The lidar – a radar-like device that bounces a laser beam, rather than radio waves, off summits, canyons, and meadows – makes initial measurements of terrain height without snow. During snow-measurement runs, the lidar measures snow height. The team subtracts the two to get snow depth. The imaging spectrometer measures the amount of light the snow reflects, important for calculating melt rates. And it reveals the nature of the surface the laser is sampling, such as snow, vegetation, or rocks, for instance.

Automated snow-measuring stations on the ground will still play a role in the measurement efforts. They dot the mountain West, providing important information about snow depth at their location and about the density of the snow. The density information feeds into the observatory's water-content calculations. But automated ground sites are set up in locations people can readily reach. They provide no information on any snow lingering between sites.

In California's Tuolumne River watershed, for instance, the highest automated measuring station is located in Dana Meadows at 9,800 feet.

"There's fully a third of the watershed above that," Gehrke says. The Tuolumne River provides irrigation water for the San Joaquin Valley and about 15 percent of San Francisco's water.

From orbit, satellites provide information on the extent of the snow pack across the entire range. But those images don't reveal snow depth or details of how patchy snowfields can be within individual basins.

Researchers are working on ways to give satellites the capability to make such distinctions, "but it's not there yet," Painter says.

During the past two years, Painter's team has improved its tools – including the aircraft it used to make the flights. It also has faced some daunting challenges, the biggest of which was providing the measurements to water managers within 24 hours of collecting the data.

The observatory visits each of its small sample of basins once a week on average.

With each flight the aircraft returns with slightly more that 700 gigabytes of data, Painter says, adding, "It takes hours just to get the data off the flight disk." Other scientists who had used similar airborne observatories to track changes in tropical-forest cover termed the notion of a 24-hour turn-around time crazy, Painter says. Yet the team managed to pull off the feat.

Beyond ready measurements for water managers, ASO data also have developed a following among scientists studying basic interactions between vegetation, soil moisture, and climate in mountains. The demand is intense enough that the team is looking for a place to build an archive of the data the observatory gathers.   

Looking back on the first two years of work, it's become clear that the ASO is an ongoing project, Painter says. "There's no going back."