s
s
Sections
Sections
Subscribe

For most of us, the reliability of the weather forecast is a minor concern, a question of whether to take along an umbrella or to when to quit watering the yard.

But to the managers of California's vast water systems, the question is more critical. A badly-timed decision can worsen the long-term effects of a drought on one hand, or allow a catastrophic flood on the other.

So on the cusp of the rainy season, managers at the Sonoma County Water Agency would very much like to know what the next few months will bring, weatherwise. With two unusually dry years behind them, supplies of water are dwindling, particularly in Lake Mendocino, which is down to a little more than a third of its maximum capacity. A solid prediction about the weather over the next couple of months would let engineers know whether to breathe a sigh of relief or tighten up even more for a bad summer next year.

Unfortunately, the forecasting tools available to them aren't any better than those used by your local TV news show. And even big picture indicators, like the cyclical El Nino phenomenon in the Southern Pacific, aren't offering any clear suggestions of what to expect this year.

"It has an equal chance to be incredibly dry, or to be incredibly wet. Or somewhere in the middle," said Don Seymour, principal water agency engineer. Scientists "don't have a strong indicator of what this year could be."

Realistically, the best weather forecasters can do for worried water managers is give an indication of patterns up to perhaps a week in the future, engineers say. And the longer the range of the forecast, the less reliable those forecasts become.

"The tools we have available to us, frankly, aren't great," agency Chief Engineer Jay Jasperse said. "We feel we're making the best of what's out there."

But that hazy picture of the weather future could be changing, thanks to new research and a network of detailed monitoring stations being built along the West Coast, starting with a new facility installed this month near Bodega Bay.

"I fully expect to see improvements over the next decade" in long range weather forecasting, said Marty Ralph, director of the Center for Western Weather and Water Extremes at the Scripps Institute of Oceanography.

In his former job as a scientist for the National Oceanic and Atmospheric Administration, Ralph led pioneering research in the Russian River watershed into understanding the effects of "atmospheric rivers," long, narrow plumes of moist air that gather along the boundaries of cold fronts and are responsible for as much as half of all precipitation that drops on the West Coast.

That research, conducted in the late 1990s with the cooperation of the Sonoma County Water Agency, led to the creation of the monitoring stations now being installed, worth around $1 million each. Three more are scheduled to come on line this year, he said.

It also raises the tantalizing possibility that weather scientists could eventually give water managers a reliable picture of rainfall days or weeks in advance, knowledge that could revolutionize the way engineers manage water systems.

"I think the future is promising," Ralph said. "I think the technology is out there, the ideas are out there; the big gap is the funding to explore these possibilities."

Here's why all this matters.

California gets most of its rain and snowfall in the space of just four months, from November through March. And much of that comes from a handful of torrential events, perhaps three to eight big storms per year.

That same lifegiving water, therefore, can create devastating floods, particularly in the Russian River watershed, possibly the most volatile river system in the west.

As a result, the state's system of reservoirs and drainage channels was built not just to store water, but to cushion the shock of major flooding. If water managers allow a reservoir to fill up to the top during the rainy season, the next storm would have nowhere to go but to flood the farms and cities downstream. It could even create enough overflow to erode earthen dams, such as the one that created Lake Mendocino, causing the dam to collapse and sending the full contents of the reservoir crashing down the river.

In effect, therefore, in the winter, just at the very moment when water managers have the best chance of collecting plentiful amounts of water, they have to force themselves to drain water off to make room for the possibility of another big storm.

Locally, that drama plays out most starkly at Lake Mendocino.

The agency's other major water source, Lake Sonoma is a much larger reservoir, built to hold about three year's worth of water and with plenty of reserve capacity to handle floods on top. Congress, however, never allocated the money to finish the dam at Lake Mendocino, leaving it too small to comfortably meet the needs of both drinking water and flood control.

The story tells itself on a graph, using a dotted line called the "storage curve." When the lake level rises above that line, the U.S. Army Corps of Engineers has to start draining off water to leave a safety margin for the next big storm.

In the summer, that line is higher since the chance of a big storm is minimal; the lake can safely store about 111,000 acre-feet, or about 36 billion gallons, of water, which can be used by the water agency to supply farms and cities and to release to preserve fish spawning habitat downstream.

In the rainy season, however, that maximum number drops nearly by half, to about 68,000 acre-feet, or 22 billion gallons. The rest of that space has to be held in reserve to control possible floods.

If the science of weather forecasting were to improve dramatically, and could confidently tell water managers that no more major flood-producing storms were on the way, the Army Corps of Engineers might be able to hold onto some of that winter water longer, giving the water agency more reserve to handle the dry summer months to come.

That goal is far down the road, engineers say, since we still know relatively little about how the atmospheric rivers form and move. Even when scientists know a river is forming, predictions of exactly where it will hit along the coast have a margin of error of hundreds of miles even just 24 hours in advance.

"At this point, we simply don't have the science down to the point where we can absolutely predict we won't have disaster on our hands" if the Corps lets the reservoir get too full in the rainy season, said Mike Dillabough, chief of the Operations and Readiness Division for the San Francisco District of the Army Corps of Engineers.

Until that day comes, Ralph said, there is still plenty that our emerging understanding of atmospheric rivers can tell us. The new monitoring stations, combined with improved rain gauges, soil moisture probes and advanced radars, can tell scientists when and where the rivers hit the coast, how fast they are moving and how much water they are dropping. That can tell engineers and emergency planners when to expect major flooding hours earlier than they previously could know.

It's not unrealistic to think that we could soon be able to do a forecast of the likely number of atmospheric rivers in any given season, the way East Coast forecasters predict the strength of a hurricane season in the Atlantic. That might not be specific enough to change the awkward balance between water storage and flood control, but it would at least give engineers some sense of what to expect in an upcoming rainy season.

And that is a luxury that the Sonoma County Water Agency's engineers, anxiously watching the level of Lake Mendocino drop, do not have today.

"That's the challenge we're facing; we don't have any miracle answer to that question," Jasperse said. "But what we are more and more engaged with top scientists who are involved in these issues."

(You can reach Staff Writer Sean Scully at 521-5313 or sean.scully@pressdemocrat.com.)