Close Mobile Menu

California’s Water System Leaks Like a Sieve—How To Save Millions of Gallons

March 1, 2016
by Glen Martin
Water spurting out of a pipe

The drought may not have caused California’s water crisis, but it’s certainly brought it to the attention of a public largely uninterested in it until government fiat made shorter showers and dead lawns de rigueur. State water demand has outstripped supply for decades. Water rights claims for the massive State Water Project and federal Central Valley Project outpace average annual supplies by at least five fold.

And that’s just for surface water. Groundwater provides 30 to 45 percent of California’s needs, and we’re pumping it so fast that massive land sinking is occurring throughout the San Joaquin Valley.

Thus, with 38 million people in the state and counting, we’ll never have “enough” water. The only thing we can do is manage what we have more efficiently. But therein lies the rub: We don’t know how much there truly is, and we don’t know how much we really use. It’s not just that groundwater is unmetered, or that the way surface water is allocated is confusing, poorly tracked and executed—and sometimes egregiously duplicitous. The infrastructure delivering our water is so antiquated and leaky that much of it is lost before it can squirt out a tap.

French drain trying to control flooding

“A tremendous amount is being wasted,” says John Gage, a founder of Sun Microsystems who’s on the advisory board of the Goldman School of Public Policy at UC Berkeley. “We haven’t even begun to accurately quantify it yet, but we know, for example, that about a million gallons of water a month leaks from the old terra cotta pipes at Cal’s botanical gardens. Leakage is a significant contributor to California’s water dilemma. The infrastructure is old, cracked, damaged by quakes. It’s a mess.”

A 2011 report from the California Public Utilities Commission estimated that 10 percent of all urban water deliveries—roughly 840,000 acre feet—are lost to leaks annually. (An acre foot is the volume of water required to flood an acre one foot deep—about 325,000 gallons, ) As noted by environmental reporter Paul Rogers in the San Jose Mercury News, that’s equivalent to a full year’s deliveries to the 5.7 million customers served by the Bay Area’s biggest water agencies: the East Bay Municipal Water District, the Hetch Hetchy Regional Water System and the Santa Clara Water District.

Still, given that California water issues are as murky and turbulent as the Yuba River during an El Nino deluge, you might wonder why a wealthy techno-whiz like Gage would even bother grappling with them. It may be as simple as this: Gage likes challenges, and water access and distribution is one of the most challenging issues emerging in the 21st Century. As part of the Human Needs Project, an initiative that provides clean water to urban areas in the developing world, he helped design and build a revolutionary water procurement and treatment system for Kibera, a sprawling and impoverished Nairobi neighborhood widely cited as Africa’s largest and toughest slum. And at a certain point in the current drought, Gage decided his home state could use a little progressive thinking on water policy.  California’s leaky water delivery system, Gage maintains, is emblematic of a larger problem. Even as we waste our water extravagantly, we have no idea how much we really have. We don’t know how much is in our aquifers, and we don’t track it accurately as it moves from snowpack to surface reservoirs and streams to end users. Until we quantify it, we manage supplies in a rational and effective way.

“Because land is subsiding all over the San Joaquin Valley, we know that aquifers are being overdrafted,” says Gage. “But people keep punching boreholes all over the place, racing for the bottom. And the guys doing the punching are resisting any effort to quantify the water because they know it would lead to regulation.”

Exacerbating the problem is the process of quantification itself. The standard technology is expensive and time-consuming to deploy. What’s needed is a system that is cheap and easy to deploy, a system that can be established on a landscape scale: not just for aquifers, but for every watershed, so surface water and groundwater can both be measured simultaneously, in real time and on an ongoing basis.

Gage ran his thoughts by Paul Sagues, an old buddy and fellow tech entrepreneur. Sagues, a Cal-educated engineer and the CEO of Xio, a firm specializing in industrial process control systems, contributes hardware and expertise to the Human Needs Project. Like Gage, he has both a professional and personal interest in water.

Leaky pipes

In fact, a certain preoccupation with water seems to run in the Sagues family. The Xio CEO has a younger brother who oversees a two-well system near the Russian River that serves about a hundred houses. In California, any water delivery entity that supplies more than a few houses constitutes a water district; there are about 8,000 such districts in the state, and virtually all of them are predicated on outmoded, often jury-rigged technology that demands a lot of human oversight and wastes both water and energy.

“Paul’s brother was constantly having to drive around checking things to make sure the pumps were running, the lines weren’t leaking and the water filtration and treatment systems were working,” says Gage. “It was basically all dependent on human oversight. And it was a pain. So he told Paul, ‘If you’re so smart, why don’t you come up with something that can help me out?’”

Sagues did have something for his little brother. He had sold an earlier company he founded, Berkeley Process Control, to General Electric for a tidy sum. Berkeley Process Control technology is used in a wide range of applications, including nuclear power plants. But GE didn’t want one of the chips Sagues’ firm had developed because it was designed for smaller industrial systems.

“It turned out to be perfect for small water systems,” says Gage. “When you put it in a little box and connect it to sensory inputs and outputs, you can automatically control water pumping, treatment and delivery, and you can monitor everything remotely. You can pump water when electricity is cheapest.  You can tell where there are leaks, and how big they are. You can monitor for chlorine levels, turbidity, various contaminants. It can run irrigation systems, supply homes, monitor water supplies—basically, it does everything required for water procurement and delivery with none of the wasted water, wasted energy and human labor associated with most current systems.”

Sagues is now involved in marketing his technology to small water districts, and has found plenty of interest; one of the systems distributes water at Squaw Valley.  

“Paul considers himself retired and essentially wants to focus on the 8,000 rural water districts in the state,” says Gage. “But I’m thinking it can be scaled easily. We can apply it to our largest water districts. We could cut slash water waste, and save a tremendous amount of electrical energy and human labor.”

Further, if conjoined with other technologies developed at Cal, the system could be used to quantify the state’s water, he says. “UC developed the sensors that are used to record and transmit snowpack information. They’re powered by solar panels, and they measure snow depth, humidity and other data points. The information is transmitted to computers and unloaded to the internet, so we now have a good idea what’s going on with the snowpack at any given time.”

Photo of John Gage
John Gage

By using advanced chips like those developed by Sagues in a similar fashion, Gage says, we can build a network that would provide an accurate picture of the state’s water supplies and usage at the stroke of a computer key. Sensors at the bottom of wells could monitor aquifer levels in real time, so you’d know the status of every groundwater basin in the state at any given moment. Sensors deployed along every stream in every watershed would allow managers to identify where water is going, who is pumping it, and when it’s being pumped. You could see how much water is coming into a city, monitor how much is being used by homes, businesses and greenbelts, how much goes into treatment plants and how much comes out. Only such meticulous measuring, says Gage, will give us the reliable information needed to allocate water equitably and efficiently.

“Right now I’m in the networking mode, sharing this information with people who work in government, the universities and the field,” Gage says, “and I think I’m making progress. I talked to (Congressman) John Garamendi recently. He lives right by the Sacramento River, and he immediately understood the significance of what we’re trying to do.”

But there’s something hobbling the application of this new technology: old technology. It goes by the acronym of SCADA (for Supervisory Control and Data Acquisition). It, too, is a system for remotely monitoring and controlling industrial processes. But it requires far more human involvement, it requires a lot of big, heavy gear and it sucks up lots of energy. And it has been around for decades.

“It uses heavy old boards, it’s in all these old buildings running HVAC systems, and it’s used by most water districts,” says Gage. “It takes a lot of equipment, electricity and human oversight to do what systems like our little chip and sensors do much more efficiently and cheaply.”

So how do we make the shift to the newer, high-tech approach? Of course that would require replacing existing infrastructure, but Gage says probably the biggest problem is the entrenched constituencies that install, service and oversee existing SCADA systems.

“The students I work with at Cal understand the new approaches, but the older guys who are running our water systems, who think in terms of SCADA, don’t really get it. They’re resisting anything new. So until they retire or are replaced, it’s going to be an uphill climb.”

Share this article