An Oroville message: As climate shifts, so will water strategies

Even when everything is going right, managing a dam is a juggling act. What the flooding this week at California’s Oroville Dam may be demonstrating is how that juggling act is growing even more complicated due to climate change.

Many factors are at play in the ongoing emergency, which has caused more than 100,000 people downstream to be evacuated. Neglect of infrastructure has played a clear and primary role – with homes being evacuated because of signs that the dam's emergency spillway is failing to safely carry even a portion of the overflow it’s licensed to handle.

By contrast, the connection to climate change is less clear. Most climate scientists say it’s hard to prove that global warming directly causes a specific extreme weather event. What they do say, though, is that warmer temperatures make the chances of such episodes occurring more likely.

In fact, the Oroville incident is raising what may be an overdue debate over how water managers can better adapt to new climate realities.

"I think we can certainly say that under a warmer climate, more of California's winter precipitation will arrive as rain, rather than snow, so that these kinds of concerns about reservoir capacity will become increasingly urgent,"  Sally Thompson, an assistant professor of civil and environmental engineering at the University of California-Berkeley, says in an email.

To her it looks “premature” to draw a direct line from climate change to Oroville’s busting-at-the-seams reservoir. But she and others say it’s not an irrelevant line of inquiry.

“The disaster unfolding at Oroville, where precipitation is more than double the average, is the latest reminder that the United States needs a climate-smart upgrade of our water management systems,” Stanford University climate scientist Noah Diffenbaugh wrote Tuesday in The New York Times.

Again, the problems at Oroville include ones that have nothing to do with shifts in climate. Environmental groups warned in 2005 that the dam’s emergency spillway needed reinforcement, for example.  

“What we're looking at is the challenge - and it's a very real and scary challenge - of managing and maintaining aging infrastructure,” Thompson says. Some policy analysts hope that the Trump administration will include funding for aging dams in a planned infrastructure spending bill. Details of that plan have yet to emerge

But Thompson adds that structural problems at dams have arisen elsewhere when heavy rains follow a period of prolonged drought, which some scientists contend is a fingerprint of climate change.

Time to bend the 'rule curves'?

Operational procedures at dams are also due for an update in the face of climate change, experts say.

Dam managers are constrained in their flexibility to adjust to changing realities posed by a warming world. The managers follow so-called “rule curves” set by the US Army Corps of Engineers that indicate when to draw down or fill up reservoirs, which take into account flood control, water supply, and endangered species requirements.

In recent years, snowpacks across the West have been lighter. Rainfall is less frequent and more intense when it comes, as the Oroville case may exemplify. The rain falling in some places used to come as snow, which would melt into the reservoir system later in the year. That rainfall also can push mountain snow into the reservoir system faster.

“There’s a push from flood-control requirements that like to keep reservoirs low at certain times of the year and a desire from people who want to have water for fish and irrigation to have reservoirs as high as possible all the time,” says David Raff, a science advisor at the US Bureau of Reclamation.

The rainstorm that hit the Oroville area was unusually warm, says David Freyberg, a hydrologist and associate professor of civil and environmental engineering at Stanford University.

“Probably the biggest surprise which would put them off the rule curve was that it was such a warm storm,” Professor Freyberg says. “They get a lot of snow in the watershed and the snow stays put, but those last storms were quite warm so it was rain.”

Competing priorities

In many cases, the rule curves are a result of negotiations between water districts and communities that rely on dams to provide their many services. The larger the dam and the more integrated into a multi-dam system, the harder it is to break from those rule curves, Freyberg says.

But those rule curves are increasingly out of whack with modern precipitation patterns and usage. Oroville Dam, the tallest in the US at 770 feet, is mainly used for water supply in southern California. But its other major purpose is controlling floods.

One function relies on a fuller reservoir, the other an emptier one.

“Operators do feel sort of constrained,” Freyberg says. “Reopening the rule curve starts a process that can take a while and become contentious. I think we’re starting to realize we’re going to need to revisit a lot of rule curves, especially in the West.”

It’s hard to pin down how much climate change is responsible for the precipitation patterns that contributed to the Oroville situation.

Precipitation in the northern Sierra Nevada mountain range is 220 percent above normal this winter, making it the wettest on record. But snowpack is just 145 percent above normal, falling short of the snowiest winter. Warmer temperatures are likely to blame for the discrepancy, says Alex Hall, an atmospheric and oceanic sciences professor at the University of California, Los Angeles.

“When we have such huge flows coming early because precipitation is falling as rain and not snow, it really does challenge the flood-control dimension of reservoir management,” Mr. Hall says.

Some argue fealty to rule curves poses problems, especially if they’re not aligned with recent precipitation patterns.

“Not only are they for an outdated period, they are actually written on graph paper with a pencil,” says John Cain, director of conservation for California flood management with American Rivers. “They weren’t even digitally generated.”

Johnny Roache, program manager for the river and reservoir operation for US Bureau of Reclamation’s Pacific Northwest region, says his region is conducting a climate change review to get better data to inform reservoir drawdown and fill dates. But he says the rule curves allow for some wiggle room.

“They aren’t etched in stone, but they are the guidance. And until they’re changed, that’s what they are. I think as part of this climate change we will probably look into that and see … do we need to draw down earlier?” Mr. Roache says.

One challenge has been estimating changes in snowmelt, Roache says. “How fast it comes off is a big variable.”

Managing the spigot for Columbia River

While the Oroville situation has pushed itself into the spotlight for its urgency, with crews struggling to keep the spillway from failing, tough water-management decisions exist even when reservoirs are smooth.

Looking out from atop the Grand Coulee Dam’s wall in northeast Washington, it’s impossible to see where Lake Roosevelt ends as it stretches 150 miles north into the Canadian Rockies that feed the Columbia River’s flow.

Here too, at a mile-wide dam that dwarfs most engineering projects ever undertaken in the US, a shift toward precipitation falling more as rain than snow makes a big difference. It means adjusting timetables for releasing water in the spring as Lake Roosevelt reservoir fills earlier in the season.

Drawing down the lake too early could flood communities downstream and leave too little water for later in the summer to supply the river system with cool water for fish, potentially running afoul of Endangered Species Act obligations. But leaving the reservoir too full risks creating spills at the dam, which in turn allows water to mix with nitrogen that can kill fish. Managers often feel hamstrung.

“You’re adjusting on the fly,” Roache says. “All of this stuff operates as a system.”

One bright spot is that forecasting capabilities are improving. That’s thanks to satellite observations and modeling, though dam managers also have called for increased funding for snowpack gauges to get an even clearer picture.

“Forecasting technology [should mean] … there’s less risk of not having water stored that could have been stored and maintaining the safety from the flood perspective that the community expects,” says Mr. Raff at the Bureau of Reclamation.

That is, if the Trump administration maintains the satellites used to collect that information. Climate research programs at the National Oceanic and Atmospheric Administration, as well the National Weather Service and NASA, are potential targets for cuts by the White House.

“My concern about the rule curves being out of date is legitimate, but a terrific mitigating factor is our ability to forecast,” says Mr. Cain of American Rivers. “And that is why it would be a huge disservice to cripple NOAA and the National Weather Service climate change science programs.”

Some of the reporting for this story, involving a visit to Grand Coulee dam, was funded by the Bill Lane Center for the American West at Stanford University.