Non-potable water reuse in San Francisco

Water reuse can be implemented at different scales
Water reuse can be implemented at different scales
Featured Faculty: Kara L. Nelson, Arpad Horvath

California’s recent severe, multi-year drought reminded citizens and policymakers alike of our dependence on unreliable water resources.  Projections for climate change suggest that these droughts may become a “new normal” across much of the western United States, creating an urgent need to develop new and more reliable sources of water. “A changing climate, combined with population growth, mean that water scarcity is the new reality in the arid West,” says Dr. Kara Nelson, a professor of Civil and Environmental Engineering at UC Berkeley. “Non-traditional sources of water need to be found to maintain the current water needs and meet future demands for our cities.”

One of the most promising opportunities is recycling wastewater (also called water reuse), and researchers from UC Berkeley have been studying the energy and greenhouse gas implications of it. (See below for detailed information about the researchers.)

Water reuse involves treating wastewater to an appropriate standard so it can be used again as non-potable or potable water rather than being discharged into an ocean, river or other water body. “Water reuse can be an efficient way to provide reliable, safe, drought-resilient, local water,” says Dr Jennifer Stokes-Draut, a Research Engineer in Civil and Environmental Engineering, who has been studying the energy and environmental impacts associated with the water and wastewater infrastructure for 15 years.

Managing water is often an energy-intensive activity. Recycled water needs to be collected, treated and redistributed to customers, processes that all require material and energy inputs and emit greenhouse gases (GHG) at the site of implementation and throughout the life cycle.

Non-potable reuse (NPR) involves reusing water for applications that do not require high quality water such as toilet flushing and irrigation. NPR has lower treatment requirements than are needed for potable reuse, but requires the construction and installation of a separate, non-potable, distribution system for the recycled water.

Water reuse at different scales

Water reuse can be implemented at different scales

“As concerns over energy reliability (such as matching supply and demand for electricity in real-time) and climate change increase, using a life-cycle approach to assess the impacts of new water supply and infrastructure is critical,” says Dr. Arpad Horvath, a professor in Civil and Environmental Engineering and a life-cycle assessment expert.

Another important consideration is the size of the system. “Non-potable water reuse can be implemented at different scales, from a single house to an entire city or region,” says Olga Kavvada, a PhD student in Civil and Environmental Engineering and the lead author of the study.

Larger, or centralized, systems are generally managed by the local water and/or wastewater utility and involve treatment facilities co-located or near a wastewater treatment plant connected to miles of distribution piping to deliver the recycled water to customers.

Smaller, decentralized plants generally serve a single building, a neighborhood, or small community, and the water is reused in and around the same buildings where it was generated. They may be managed by a utility, a property owner, or other entity, depending on local practice.

Larger systems benefit from economies of scale because the treatment takes place in large facilities that are generally more efficient. The tradeoff is that larger scale systems require more infrastructure and, often, more pumping for both the wastewater being collected and the treated water being distributed, as they are generally located further from the customer.

“This tradeoff is generally understood but has been poorly analyzed,” says Olga Kavvada, “because it depends on many site-specific factors, such as land use, water demands, population density, and topography. Spatial analysis along with a life-cycle assessment (LCA) of the required infrastructure is needed to identify options that increase water resiliency while minimizing environmental impacts. Using LCA allows us to get a holistic understanding of the environmental impacts of the system’s life-cycle from material extraction to operation.”

spatial analysis and lifecycle methodology
This study combines spatial analysis and a life-cycle assessment methodology to estimate environmental impacts of water reuse

How is San Francisco promoting non-potable reuse?

The San Francisco Public Utilities Commission, the city’s water provider, has been a leader for exploring and adopting innovative approaches to enhance their water supply.

Through conservation, the city has reduced per capita water consumption to one of the lowest reported levels in the state.

To further reduce water use, the city has championed the practice of NPR in buildings, with recycled water used for toilet flushing and landscape irrigation. As of November 1, 2016, all new large buildings in San Francisco are required to provide an alternative water supply, which in many cases will involve building or district-scale NPR.

Where does decentralized reuse make most sense in San Francisco?

“In our recent study, we combine spatial analysis and LCA methodology to estimate environmental impacts of decentralized versus centralized NPR water reuse for toilet flushing, using San Francisco as a case study,” says Olga Kavvada.

The study reveals the tradeoffs between having a smaller conveyance system for decentralized NPR versus the efficiency in the treatment process as the amount of treated water increases.

The figure below visualizes the results of the analysis. Green cells denote places where implementing smaller-scale reuse makes the most sense. Red indicates that centralized reuse will make more sense at the scales assessed.

“The colors identify the minimum system scale where decentralized reuse has lower impacts than centralized reuse,” says Olga Kavvada. If decentralized NPR were to be implemented at a scale equal to or greater than the size indicated (in units of m3/day) in the specific area of the city shown on the grid, it will be more energy efficient than connecting to reuse from the centralized treatment plant.

“At the minimum scale and larger, the impact of water conveyance from the distant wastewater treatment plant outweighs the treatment economies of scale,” says Olga Kavvada, “making a small decentralized system more efficient overall.”

decentralized vs. centralized scenario

Minimum facility size for each grid cell for decentralized reuse to be more efficient than a centralized reuse scenario for (1) “completely dispersed” and (2) “completely connected” recycled water pipe networks. Lifecycle energy intensity for (a) current MBR performance and (b) future scenario in which the MBR treatment has gained 20% operational efficiency.

Why does this matter?

“Water infrastructure, once constructed, lasts for decades,” says Dr. Thomas Hendrickson, a technical specialist at ICF in San Francisco, formerly of UC Berkeley. “The systems we implement today will affect energy consumption and GHG emissions for years to come. These long-term effects must be explicitly considered in our planning and design choices.”

All water infrastructure is spatially sensitive. Optimal solutions could vary depending on location, scale, number of facilities and topography.

Because centralized wastewater treatment plants are typically located at low elevations (to ensure gravity flow of sewage) and at the outskirts of the cities, water reuse systems are particularly sensitive to energy needed to reach the customer.

Although small systems are less efficient in treatment efficiency, they benefit from being closer to the point of use, avoiding the conveyance energy and GHG penalty.

“By carefully assessing the spatial conditions in which NPR exists, we can identify optimal solutions that minimize the environmental impacts now and in the future,” says Olga Kavvada.

Centralized infrastructure, though perceived as more reliable and efficient, presents barriers for NPR, because large-scale dual-distribution systems can be costly and disruptive to implement in dense urban areas.

“Decentralized infrastructure allows for a flexible and incremental approach to system expansion in the context of uncertain future growth patterns,” says Dr. William Eisenstein, a policy expert and member of the study team. “Water recycling can also increase community engagement about the value of water and provision of this resource in the future.”

Relevant Work

The methodology and the results of the study can be found in an Environmental Science and Technology publication, Assessing Location and Scale of Urban Non-Potable Water Reuse Systems for Life-Cycle Energy Consumption and Greenhouse Gas Emissions. (DOI: 10.1021/acs.est.6b02386)

For further information, see:


Funding for this research was provided by the National Science Foundation Engineering Research Center, Reinventing the Nation’s Urban Water Infrastructure (ReNUWIt). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the researchers and do not necessarily reflect the views of the NSF.

Arpad Horvath and Thomas Hendrickson gratefully acknowledge the financial support of Henry H. Wheeler, Jr., and dedicate this research to his memory.

About the Researchers

Olga Kavvada is a PhD Student in Civil and Environmental Engineering at UC Berkeley. Her work involves systems-level analysis to improve the usage of energy and water resources under the constraints of climate change.

Dr. Kara Nelson is a professor of Civil and Environmental Engineering at UC Berkeley. Her teaching and research address innovative strategies to increase the sustainability of urban water infrastructure around the world. She leads the engineering research thrust in ReNUWIt.

Dr. Jennifer Stokes-Draut is a Research Engineer in Civil and Environmental Engineering at UC Berkeley. She studies the life-cycle energy and environmental impacts and tradeoffs associated with water and wastewater infrastructure under current and uncertain future conditions.

Dr. Arpad Horvath is a professor of Civil and Environmental Engineering at UC Berkeley. His teaching and research involves life-cycle environmental and economic assessment of products, processes, and services to answer important questions about civil infrastructure systems and the built environment.

Dr. William A. Eisenstein leads the urban systems and institutions research thrust in ReNUWIt.  He is the executive director of the Center for Resource Efficient Communities at UC-Berkeley, which focuses on strategies to reduce the water, energy, and greenhouse-gas intensity of cities.

Dr. Thomas P. Hendrickson is a technical specialist at ICF in San Francisco. He specializes in environmental mitigation and climate adaptation assessments of complex infrastructure systems, with a focus on identifying resilience solutions for water and wastewater systems.


Published 12/21/2016