Water is an economically scarce, environmentally sensitive resource. Limited options for new water sources, increasing regulations on quality, and significant growth in demand are challenges faced by those who plan the future needs of our cities.

Accurately forecasting future water demand is critical to economies, communities, and ecosystems, especially where available water must be allocated among competing uses. How customers are using water has a great influence over how much will be needed in the future. Communities must consider numerous factors that affect water use, including economic development, weather variability, shifting urban and agricultural needs, residential growth patterns, and improvements in water-use efficiency. Identifying and measuring trends can help managers make better decisions by anticipating capital investment needs and allocating resources better—thereby maximizing long-term value and reliability of resources.

Traditional forecasting of water needs remains simplistic—typically tying future needs exclusively to population growth. Relying only on population provides limited data to planners and fails to capture the underlying factors driving per capita water use. The shifting industrial landscapes of communities, changing residential growth patterns, and improvements in water use efficiency result in changes in per capita water use, while traditional forecasts assume a constant rate of use over time.

Thus, two communities may have similar per capita totals, but based on demographic trends, area-specific industries, residential concentration, and numerous other determinants, the amount of water they each will need in the future may differ. Traditional population-based approaches cannot factor in these underlying influences and therefore don't lead to an understanding of how water use has changed and will change in the future.

More Sophisticated Forecasting

Advanced methods of water demand forecasting recognize that changes in demand stem from many factors. Estimates of future demand can be developed that take into account the shifting conditions of the future, whether they are socioeconomic trends, climatic influences and seasonal variations, changing water and sewer rates, or increasing water-use efficiency.

Factors that affect the average rate of water use can be projected and used by planners to develop alternative scenarios and to assess the sensitivity of future water demand to changing conditions. By considering the variation in each of these influencing factors, planners can develop probabilistic water demand forecasts that can be used in risk-based demand simulations. More comprehensive, meaningful water demand forecasts lead to more accurate, reliable, and cost-effective decisions by planners.

Critical issues are confronting urban water systems that are faced with population and economic development pressures. Many Americans are moving to and doing business in areas that have environmental, financial, or political barriers for water supply augmentation— such as booming communities in the southern and southwestern United States. Other geographical regions not customarily viewed as water-scarce face large potential investments to rehabilitate and expand aging systems to serve future demands for water.

Several communities across the country have embraced the power of a more sophisticated approach to water demand forecasting as a means for better planning. The city of San Diego, which imports the majority of its water, has developed a progressive water demand forecasting program to anticipate the capacity and construction needs of future water supply projects that fit its unique circumstances.

Sustaining Growth

San Diego is a large, semi-arid city that serves more than 1.2 million people across more than 130 square miles of developed land. It imports 90% of its potable water (about 200 mgd), which is transported to the city from the Colorado River and from northern in-state sources. Unlike typical water providers, San Diego acts as an agency that charges its customers for water at the retail level, which can be affected by the prices it pays for imported supplies and the amount of water it may need to purchase.

San Diego began using advanced water demand forecasting in the 1980s, when Planning and Management Consultants Ltd. (now part of Camp, Dresser & McKee Inc.) developed a unique set of water demand models for the city service area. Since then, water demand models have been estimated for the San Diego County Water Authority that apply to the city of San Diego service area. These models are updated as new water use and demographic information becomes available.

Recently, changing residential growth patterns and a large capital improvement program have introduced intricacies to the city's water management processes. As a member of the California Urban Water Conservation Council (CUWCC), which monitors water use efficiency and provides conservation guidance throughout California, the city provides periodic updates of its future water conservation savings. San Diego also updates its future water demands in its urban water management plan for the California Department of Water Resources every five years. To comply with these reporting requirements and to meet its own internal planning needs, the city's water demand is updated as new models and model inputs become available.

In conjunction with the San Diego County Water Authority—the wholesale, regional group that provides water for the city—the city developed a 30-year water demand forecast with appropriate inputs calibrated for the makeup of the city. Forecasts were estimated for the single-family, multifamily, nonresidential, and wholesale sectors for 35 major pressure zones within the city, organized geographically. The water use models developed for each sector included model variables that account for typical monthly variation in use, climatic deviations from average monthly weather conditions, socioeconomic characteristics, and the price of water and sewer service.

Average rates of water use per household varied significantly among pressure zones as characteristics such as housing density, median household income, and persons per household varied throughout the designated sectors. Each pressure zone had a unique mingling of single-family and multifamily housing and different projections of the growth rates for these two residential sectors. Average nonresidential water use varied among major zones as the combination of retail, wholesale, industrial, and government employment varied from one pressure zone to another. As with the housing projections for the residential sectors, projections of future employment by the major employment groups varied among the pressure zones. Thus, the water demands of each sector reflect distinct patterns of water use and varying demographic trends among its users.

Considering all determinants, the city estimated water demand for each major zone based on projections of housing, housing density, persons per household, income, employment, price of water and sewer service, and climatic factors over several decades. These demand forecasts provide San Diego with full use of regional planning data, greater understanding of spatial trends as they relate to the community's growth, and future needs in dry/hot demand scenarios. Better equipped, the city can make assumptions about economic growth, water pricing, and other major factors influencing usage rates and use these forecasts when determining the cost of water supply options, facility sizes, and estimated annual costs of imported water purchases.

Currently, long-range estimates of water demand are being used by the San Diego Water Department to develop various reports (such as its strategic water supply plan, its urban water management plan, and its raw and treated water plan), and to plan for future improvements and optimization of the water distribution system.

Formula For the Future

Water demand forecasting provides key inputs for potentially risky decisions concerning large capital investments. If done properly, a forecast of water demand reveals the effects of those factors that cause demand to change over time. The additional information afforded by more advanced approaches can help define better alternatives for meeting future needs and highlight important trade-offs among strategies for ensuring water supply reliability. Traditional forecasting may be insufficient for pinpointing the timing and scale of system expansion and for justifying those needs to customers and governing bodies.

Decision-makers know that uncertainty drives the need to plan for the future. Although San Diego's circumstances are unique, other communities and water providers who experience pressures from rapid growth and potentially large capital investments can benefit from this methodology. By understanding and incorporating the many factors that influence water demand, a good water demand forecast translates uncertainty into information that can help communities make smart and informed choices to ensure reliable water supplies.

Davis is a senior economist and Kiefer is a principal economist for Carbondale, Ill-based CDM.