skip to Main Content
VIEWPOINTS
The PPIC Blog
A Little Girl Getting A Glass Of Water

Recycled Drinking Water: The Next Frontier

Lori Pottinger October 17, 2016

California is poised to become an early adopter of the direct reuse of purified wastewater as a source of drinking water. The State Water Board recently released a report for public comment that indicates it is feasible to regulate direct potable reuse to produce safe and reliable drinking water (comments are due by noon on October 25, 2016). We talked to David Sedlak—one of the 12 experts who worked on the report and a member of the PPIC Water Policy Center research network—about this potential new water source.

Public Policy Institute of California: How can treated wastewater be used?

David Sedlak: There are two main ways we reuse municipal wastewater. The first is referred to as non-potable reuse, which is the practice of taking water from conventional sewage treatment plants and subjecting it to a little more treatment before using it for landscape or agricultural irrigation or for an industrial use, like cooling towers or boilers. The second is to put it through a conventional sewage treatment plant and then through an advanced treatment plant, and reintroduce it back into the drinking water supply. This practice is referred to as potable water reuse.

We’ve spent the past three decades making lots of investments in non-potable reuse projects and in many cases, they turned out to be more expensive than we expected and less helpful in terms of preventing water shortages. In the early days of non-potable projects, we got all of the low-hanging fruit—using the water in places close to the treatment plant that needed it, such as golf courses or oil refineries. Because the users were close to the water source, the piping systems didn’t have to be very extensive. But as we tried to build more projects, the distances got longer and the projects got more expensive.

Potable water reuse holds a lot of promise because if you can make it clean enough to drink, you can use the existing water distribution system. In California, about half of the water use in cities is indoors. Hypothetically, there is a potential to recycle all the water used indoors—though you lose about 20 percent of it in the treatment process when you employ reverse osmosis membranes. So the upper boundary for potable water reuse might allow us to expand our urban water supply by about 40 percent.

PPIC: Where are we in terms of developing more potable projects?

DS: In California, all potable reuse systems built to date involve putting the wastewater through treatment, then putting it into underground aquifers until it is needed. This time spent in the natural environment serves to break the direct connection between wastewater and drinking water. This practice is called indirect potable reuse. What’s being discussed now is the possibility that we might skip that step. The reason people have become more interested in this approach is that not every city has a good groundwater aquifer near their water recycling plant. In the case of Los Angeles, for example, they would have to build an expensive pipeline to move treated water to valley aquifers. The places where direct potable reuse is getting the closest scrutiny are San Diego, Los Angeles, and San Jose.

Direct potable reuse is already happening in Texas—three projects have been built and a fourth is in the planning stages. The main impediment here in California is that the state has never written a permit for such a facility—no one ever asked for one before. The facilities in Texas got people thinking about the feasibility of doing it here.

PPIC: What factors affect the cost of this water source?

DS: Direct reuse is not necessarily more expensive than indirect potable projects. Engineers looking into direct potable reuse are considering additional treatment steps to reduce the risk that the failure of one or more steps in the process could cause a public health problem; these additional steps would increase the cost. But that is likely to be offset by the reduced costs of moving water, as it will be piped through the normal system. At this point we don’t see any major engineering challenges in direct potable reuse that we haven’t already seen in indirect reuse projects. The panel looked at various complicating aspects but none is a deal breaker at this point. So while the cost will vary from project to project, it looks like it will still be considerably less expensive than seawater desalination and many other alternative sources.

The bigger complicating factors are not engineering ones—these technologies have been pretty well tested in Texas and in the existing plants in California. It’s whether the public embraces direct potable reuse and if the state’s regulators feel comfortable permitting and endorsing it.

Learn more

Read about the State Water Board’s process on direct potable reuse (including how to comment on the report)
Read California’s Water: Water for Cities (from California’s Water briefing kit, October 2016)
Read “Water Management’s High-Tech Future” (PPIC Blog, September 3, 2015)

News and analysis of California policy issues from PPIC

Subscribe

by Email

by RSS

TWEETS @PPICnotes



Follow Us

Back To Top