Abstract
AbstractWater demand forecasting is an important tool for integrated water management. Long-term demand predictions include inherent uncertainty, but few studies have investigated empirical methods to forecast water demand that include uncertainty from many driving factors or evaluate how changes in demand affect linked systems such as wastewater conveyance and treatment. This study presents a novel approach for long-term urban water demand forecasting that (1) integrates uncertainty in technology, behavior, and climate; and (2) estimates effects of demand changes on wastewater management systems. We demonstrate this using a case study of 350 urban water supply agencies and more than 300 wastewater treatment systems in California serving 31 million people. For the period from 2020 to 2030, we used modeling with Monte Carlo simulations to estimate future scenarios of urban water demand given changes in efficiency, climate, water rates, and population. Effects of demand changes on wastewater management were estimated using network modeling and novel evaluation metrics. In modeled scenarios, future per capita water demand in California cities decreases by 8% to 12% depending on the included factors, while total volumetric demand declines by up to 11%. Efficiency changes had the largest impact. Wastewater agencies experience up to a 10% reduction in wastewater influent flow, but differences exist across regions and facilities. Aligning future water supply investments with integrated demand forecasts could forgo as much as $860 million in investments for new water supply capacity, but tradeoffs exist if such new capacity is important for future supply reliability with severe drought. Results demonstrate the value of incorporating social, technological, and climate factors in demand forecasting through stochastic approaches.